Using Subaqueous Soils Data to Manage Coastal Ecosystems: Implications for Bivalve Recruitment, Aquaculture, and Restoration

University of Rhode Island DigitalCommons@URI

Open Access Dissertations

2016

Using Subaqueous Soils Data to Manage Coastal Ecosystems: Implications for Bivalve Recruitment, Aquaculture, and Restoration

Brett Matthew Still University of Rhode Island, [email protected]

Follow this and additional works at: https://digitalcommons.uri.edu/oa_diss

Recommended Citation Still, Brett Matthew, "Using Subaqueous Soils Data to Manage Coastal Ecosystems: Implications for Bivalve Recruitment, Aquaculture, and Restoration" (2016). Open Access Dissertations. Paper 437. https://digitalcommons.uri.edu/oa_diss/437

This Dissertation is brought to you for free and open access by DigitalCommons@URI. It has been accepted for inclusion in Open Access Dissertations by an authorized administrator of DigitalCommons@URI. For more information, please contact [email protected]. USING SUBAQUEOUS SOILS DATA TO MANAGE COASTAL ECOSYSTEMS:

IMPLICATIONS FOR BIVALVE RECRUITMENT, AQUACULTURE, AND

RESTORATION

BRETT MATTHEW STILL

A DISSERTATION SUBMITTED IN PARTIAL FULFILLMENT OF THE

REQUIREMENTS FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY

BIOLOGICAL AND ENVIRONMENTAL SCIENCES

UNIVERSITY OF RHODE ISLAND

2016

DOCTOR OF PHILOSOPHY DISSERTATION

BRETT MATTHEW STILL

APPROVED:

Dissertation Committee:

Major Professor Mark Stolt

Jose Amador

Candace Oviatt

Nasser H. Zawia DEAN OF THE GRADUATE SCHOOL

UNIVERSITY OF RHODE ISLAND 2016

ABSTRACT

Coastal ecosystems continue to be negatively impacted by increased development and anthropogenic inputs resulting in nutrient enrichment, reduced water quality, loss of seagrasses, sedimentation, and coastal acidification. These stressors, along with historic over harvest and disease, have resulted in the collapse of commercial oyster fisheries in many estuaries worldwide. Expansion of oyster aquaculture has reversed this trend, creating a growing market for oysters as a food resource. This growth however, is being constrained by a number of issues, particularly access to the coastal zone, and identification of productive locations for aquaculture lease development in areas free of conflicting uses.

Coastal ecosystem managers have identified a critical need for a support tool that can guide development of bivalve aquaculture, and site selection for restoration while avoiding user conflicts in the coastal zone. To evaluate the potential of subaqueous soil maps as a tool for managing aquaculture development and restoration site selection, my dissertation focused on three areas of research. Firstly, I conducted in- situ sampling of surface soil pH within several mapped soil types found within coastal lagoons and embayments, to characterize pH variability and determine if coastal acidification may influence bivalve recruitment. Secondly, I identified the soil properties that related to oyster productivity for on-the-bottom aquaculture systems by conducting oyster growth trials for dominant soil landscapes within both coastal lagoons and embayments. Lastly, I developed a decision support tool that combined the results from the previous experiments along with conflicting use information to quantify the spatial extent of conflicting uses and potential development of bivalve

aquaculture within the coastal salt pond region using standardized subaqueous soil maps.

I used a hydropedological approach to assess the spatial variability of coastal acidification within two coastal lagoons and embayments in Rhode Island by measuring oyster shell dissolution, pH within the water column, and pore water pH within the upper 5 cm of the underlying subaqueous soils. Sampling and monitoring sites were stratified based on submerged soil-landscape types mapped at the Great

Group level as Haplowassents, Sulfiwassents, and Psammowassents. Using a linear mixed modeling approach, we found that pore water pH varied significantly among soils and with depth. Median pore water pH was significantly greater in sandy, low organic matter content Psammowassents (7.97) than the finer textured, higher soil organic matter content Sulfiwassents (7.35), and the Haplowassents (6.57) that receive groundwater discharge from the surrounding subaerial soils. Juvenile calcifying organisms can experience acidic stress at pH values below 7.6; thus, current pH values within the upper few centimeters of Sulfiwassents and Haplowassents may be low enough to impact recently set juvenile calcifying organisms inhabiting these soils.

Consequently, mean shell loss during a 4-wk period was significantly greater in the

Sulfiwassents (1.54%) than the Psammowassents (0.96%), with the greatest shell loss

(18.62%) in one of our Haplowassent sites with groundwater discharge.

I compiled growth rate and survival data from growth trials conducted with juvenile eastern oysters (Crassostrea virginica) in dominant subaqueous soil types over five growing seasons within Rhode Island coastal estuaries (two years of these growth trials were conducted by a former graduate student in the Laboratory of

Pedology and Soil Environmental Science using the same study design). Using a linear mixed modeling statistical approach, I found that oysters grown in sandy firm substrates (Haplowassents and Psammowassents) showed increased growth rates and survival when compared to oysters grown in silty substrates with low bearing capacity

(Sulfiwassents). These results suggest that substrate type may assist in identifying portions within estuaries that exhibit greater seston flux without having to conduct extensive hydrodynamic modeling. Sites with increased seston flux have been shown to positively influence growth rate.

Using the results from the previous studies, I developed a GIS-based support tool to couple subaqueous soils data with spatial data of non-compatible uses. I found that between 43% to 70% of the coastal salt pond region represents non-compatible uses for aquaculture development, including boating and navigation, submerged aquatic vegetation, and recreational shellfishing. Of the remaining available area, soil landscapes that can support productive on-bottom culture ranges from 2% - 34%, depending on the coastal salt pond. Currently, 2% of the coastal salt ponds are leased for aquaculture, leaving 3% (143 acres) available for lease development, given current regulations.

Our research suggests that subaqueous soil maps are a good way to stratify estuarine substrates to identify preferred soil landscapes for on-the-bottom oyster aquaculture development and restoration site selection, as well as areas prone to acidification. As the extent of subaqueous soil survey continues to expand along the

Atlantic coast, subaqueous maps will increasingly be available as a planning tool to guide use and management of the coastal zone.

ACKNOWLEDGMENTS

This research was funded by research grants from the Rhode Island Agricultural

Experiment Station, and the Nature Conservancy’s Global Marine Program.

Additional support was provided by the Department of Natural Resources Science at

URI, Watershed Watch, and the Natural Resources Conservation Service of Rhode

Island. I am grateful for the support.

I would like to thank my major professor Dr. Mark Stolt for his vision and dedication to advancing the discipline of soil science into new areas of research and management. Mark, your support and encouragement throughout this process was instrumental. I would also like to thank my committee Dr. Jose Amador, Dr. Michael

Rice, Dr. Candace Oviatt, and Dr. Arthur Gold. Your guidance and expertise provided me with the inspiration to push the concepts and ideas of soil science into the coastal zone, and to explore the connections between the science and management of coastal ecology and aquaculture.

The fieldwork for this project would not have been possible without the help of numerous individuals that provided support and a helping hand. I am grateful to Jim

Turenne and Maggie Payne from the Natural Resources Conservation Service for their knowledge and assistance in the field. To the former and current graduate students in the Laboratory of Pedology and Soil Environmental Science, I appreciate the comradery and sense of community in the lab. I benefited greatly from your assistance, and your accomplishments, which helped pave the way forward. I would also like to thank the Coastal Fellowship Program and the Fellows that assisted with the fieldwork including Ashley Merlino, Ariana Pucci, Kristopher Plante, Maryanne

Diffin, Adiza Ama Owusu, and Mason Garfield. It was a joy to work with all of you.

You not only assisted in the lab and the field, but also supported me on my professional journey.

I also owe a debt of gratitude to my parents, who instilled in me a strong work ethic, and an understanding of the value of education. Your support and guidance this journey has been so valuable to me.

Lastly, I would like to thank my wife, Pamela. Coming to URI put me on a path to meet my soulmate. Without your kindness and love, as well as support and encouragement, this process would have been much more difficult. Thank you.

PREFACE

This Dissertation was prepared in manuscript format as specified by the

University of Rhode Island Graduate School guidelines. Manuscript 1 entitled

“Subaqueous Soils and Coastal Acidification: A Hydropedology Perspective with

Implications for Calcifying Organisms” was published in the Soil Science Society of

America Journal in March 2015. Manuscript 2 entitled “Oyster Growth and Survival

Across Subaqueous Soil Landscapes and Estuaries in Rhode Island” is formatted for publication in the Journal of Shellfish Research. Manuscript 3 entitled “Spatial

Planning for Oyster Aquaculture: Application of Subaqueous Soil Maps” is formatted for publication in the Soil Science Society of America Journal.

vii

TABLE OF CONTENTS

ABSTRACT ...... ii

ACKNOWLEDGMENTS ...... v

PREFACE ...... vii

TABLE OF CONTENTS ...... viii

LIST OF TABLES ...... xi

LIST OF FIGURES ...... xv

MANUSCRIPT – I: SUBAQUEOUS SOILS AND COASTAL ACIDIFICATION: A HYDROPEDOLOGY PERSPECTIVE WITH IMPLICATIONS FOR CALCIFYING ORGANISMS ...... 1

ABSTRACT ...... 2 INTRODUCTION ...... 3 METHODOLOGY ...... 8 Study Sites ...... 8 Soil Characterization...... 9 pH Sampling ...... 11 Shell Dissolution ...... 12 Statistical Analysis ...... 13 RESULTS AND DISCUSSION ...... 14 Pore Water and Water Column pH ...... 14 Shell Dissolution ...... 17 Pore Water pH and Soil Characteristics ...... 18 Implications ...... 19 ACKNOWLEDGEMENTS ...... 20 CITATIONS ...... 21 TABLES ...... 27 FIGURES ...... 36 MANUSCRIPT – 2: OYSTER GROWTH AND SURVIVAL ACROSS SUBAQUEOUS SOIL LANDSCAPES AND ESTUARIES IN RHODE ISLAND ...... 40

viii

ABSTRACT ...... 41 INTRODUCTION ...... 42 METHODOLOGY ...... 46 Study sites ...... 46 Oyster Growth Trials ...... 47 Water Quality ...... 49 Soil Characterization...... 49 Statistical analysis ...... 50 RESULTS ...... 52 Water Quality Characteristics ...... 52 Oyster Growth...... 53 Percent Market Size ...... 55 Oyster Survival ...... 56 DISCUSSION ...... 57 ACKNOWLEDGMENTS ...... 62 CITATIONS ...... 63 TABLES ...... 71 FIGURES ...... 85 MANUSCRIPT 3 - SPATIAL PLANNING FOR OYSTER AQUACULTURE: APPLICATION OF SUBAQUEOUS SOIL MAPS ...... 94

ABSTRACT ...... 95 INTRODUCTION ...... 97 METHODOLOGY ...... 102 Study Area ...... 102 Subaqueous Soils ...... 103 Geographic Information Systems ...... 105 RESULTS ...... 107 Subaqueous Soils and Aquaculture Lease areas ...... 107 Aquaculture Restriction Zone ...... 108 ARZ and Available Subaqueous Soils ...... 109 DISCUSSION ...... 110

Summary and Conclusions ...... 116 ACKNOWLEDGMENTS ...... 118 CITATIONS ...... 119 TABLES ...... 125 FIGURES ...... 131

LIST OF TABLES

TABLE PAGE

Table 1.3. Kruskal–Wallis ANOVA on ranks (H, α = 0.05) for pore water pH samples pooled by Great Group………………………………………………………………..30

Table 1.6. Kruskal–Wallis ANOVA on ranks (H, α = 0.05) for mean shell loss percentage pooled by Great Group…………………………………………………...33

Table 2.2. Descriptive statistics for pooled water quality parameters collected with a YSI multi-meter across the growing seasons (2008 – 2009, 2011 – 2013). Kruskal- Wallis ANOVA on ranks (H) and Dunn’s multiple comparison tests were run on each parameter across the waterbodies. Letters represent statistical differences across the waterbodies for each parameter (α = 0.05). Dashed line separates coastal ponds from embayments……………………………………...…………………………………...73

Table 2.3. Chl-a data collected during the oyster growth trials (2008 – 2009, 2011 – 2013). Significant differences among locations were determined using Kruskal-Wallis ANOVA on Ranks (H). Letters represent significant differences when comparing median chlorophyll across water bodies using Dunn’s multiple comparisons (H = 140.145, df = 3, P = <0.001, α = 0.05). Dashed line separates coastal ponds from embayments……………………………….………………………………………….74

Table 2.4. TSS data collected during the oyster growth trials (2008 – 2009, 2011 – 2013). Significant differences among locations were determined using Kruskal- Wallis ANOVA on Ranks (H). Letters represent significant differences when comparing median TSS across water bodies using Dunn’s multiple comparisons (H = 9.855, df = 3, P = 0.020, α = 0.05). Dashed line separates coastal ponds from embayments………………………….…………………………..…………………...75

Table 2.5. Environmental parameters considered ideal for juvenile – adult stage eastern oysters from the literature…………………………………………………….76

xii

Table 2.9. Percent market size (% >76 mm) for oysters grown in standard aquatrays after two growing seasons. The numbers in parentheses for seed 3 represent oysters that were grown directly on the bottom during the 2nd growth year in 2013……………………………………………………………...……………………80

Table 2.12. Percent survival at the end of the growing season for each site and seed source. Percent survival was not determined at the end of the 2008 growing season…………………………………………………………………………………83

Table 3.1. Current bivalve aquaculture production value and lease acreage within the New England region…………………………………………………………………126

Table 3.2. Geographic Information Systems data developed for analysis of space use with in the Rhode Island coastal salt pond region. These data were aggregated into an aquaculture restriction zone to assist with aquaculture development planning……..127

Table 3.3. The total acreage and relative percent of subaqueous soil Great Groups within the coastal ponds. The subaqueous soils data were aggregated at the Great Group level as Haplowassents, Psammowassents, and Sulfiwassents. The total percent data represents the percent of the total acreage occupied by each of the subaqueous soil Great Groups……………………………………………………………………128

xiii

Table 3.4. Existing aquaculture lease acreage in the coastal salt pond region……...129

Table 3.5. The total acreage of incompatible uses that comprises the aquaculture restriction zone within each coastal salt pond. The data were compiled from existing RIGIS datasets, SMP, and data we derived for this study…………………………..130

xiv

LIST OF FIGURES

FIGURE PAGE

Figure 1.1. Map of study sites in Rhode Island. Inset maps represent soil sampling locations in Greenwich Bay (GB), Wickford Harbor (WH), Quonochontaug Pond (QP), and Ninigret Pond (NP). We sampled a variety of subaqueous soil landforms within each system, including bay floor (BF), depositional shoreline (DS), lagoon bottom (LB), mainland beach (MB), mainland cove (MC), platform (P), and washover fan flat (WFF). Sites sampled with groundwater discharge were also identified as GW1 and GW2………………………………………………………………………..37

Figure 1.2. Mean pore water pH depth profiles for the upper 5 cm of sampled surface horizons (Haplowassent, Sulfiwassent, and Psammowassent profiles are pooled by Great Group classification); N is the number of pH samples taken at each depth. Kruskal–Wallis ANOVA on ranks (H, α = 0.05) indicates significant differences between median pH values of the Great Groups. Vertical dashed line represents potential physiological stress zone below which juvenile bivalves and other calcifying organisms may be negatively impacted. Error bars represent SE of the mean……….38

Figure 1.3. Shell loss percentage for the soil vs. water column treatment across sites pooled by Great Group. Kruskal–Wallis ANOVA on ranks (H, α = 0.05) indicates significant differences across Great Groups. Error bars represent SE of the mean…..39

Figure 2.1. Map of study sites in Rhode Island. Inset maps represent locations of oyster growth trials in Greenwich Bay, Wickford Harbor, Quonochontaug Pond and Ninigret Pond…………………………………………………………………………86

Figure 2.2. Pooled mean growth rate (mm/day) across each year of the growth trials. Overall, the mean oyster growth rate was significantly different across years. [one-way ANOVA, F (4,42) = 8.384, P = <0.001]. Letters represent differences across years (Fishers LSD)…………….…………………………………………………………...87

Figure 2.3. Pooled mean growth rate (mm/day-1) in the coastal ponds (Ninigret and Quonochontaug Pond) vs. embayments (Wickford Harbor and Greenwich Bay). The growth rate in the embayments was significantly greater than the coastal ponds [one- way ANOVA, F(1,45) = 13.300, P = <0.001]………………………………………..88

Figure 2.4. Pooled mean growth rate (mm/day-1) during the first season vs. second season of the growth trials. The growth rate in the first season was significantly greater than the second season [one-way ANOVA, F (1,45) = 25.758, P = <0.001]………...89

Figure 2.5. Significant soil related covariates that we investigated within the three candidate oyster growth linear mixed models. Each covariate was included in separate models due to high collinearity across these covariates. Gray bands and error bars represent 95% confidence intervals. A) Percent Sand candidate model showing the predicted average growth rate across A-horizon percent sand. B) Percent Organic Matter candidate model showing the predicted average growth rate across A-horizon percent organic matter. C) Soil Taxonomy Great Group candidate model showing the predicted average growth rate across soil taxonomy at the great group level……...... 90

Figure 2.6. Significant soil related covariates that we investigated within the two candidate percent market size generalized linear mixed models. Each soil covariate was included in separate models due to high collinearity across these covariates. Gray bands and error bars represent 95% confidence intervals. A) Percent Organic Matter candidate model showing the predicted percent market size across percent A-horizon soil organic matter, B) Soil Taxonomy: Great Group candidate model showing the predicted percent market size across soil taxonomy at the great group level……...…91

Figure 2.7. Covariates that we investigated within the percent survival generalized linear mixed model. Gray bands and error bars represent 95% confidence intervals. A) Soil Taxonomy Great Group covariate showing the predicted percent survival across soil taxonomy at the great group level. B) Year covariate showing the predicted percent survival across oyster growth trial years. C) Oyster start size covariate showing the predicted percent survival across avg. oyster start size at the beginning of the growth trials. D) Oyster growth rate (mm/day) covariate showing the predicted percent survival across observed oyster growth rates during the growth trials………92

Figure 2.8. Subaqueous soil interpretation for oyster aquaculture development using on-the-bottom methods based on results of growth trials. Psammowassent and Haplowassent soils depicted in yellow and pink represent areas that may provide increased growth rates relative to the Sulfiwassent soils depicted in red…………….93

Figure 3.1. Site map of southern Rhode Island coastal salt pond region. Labeled coastal salt ponds that currently support bivalve aquaculture leases………………..132

Figure 3.2 The pie charts represent the acreage and percent of each coastal pond that is occupied by the aquaculture restriction zone (ARZ) and potentially available areas open to bivalve aquaculture development. The coastal ponds with the largest washover fan landscapes Ninigret, Winnapaug and Quonochontaug ponds respectively have the greatest acreage of Psammowassents available for bivalve aquaculture development…………………..133

xvi

Figure 3.3. Map depicts subaqueous soils classified at the great group level within Winnapaug Pond. Areas likely to not support the development of bivalve aquaculture due to incompatible use (eelgrass, navigation/moorings, shellfish spawner sanctuaries, recreational shellfishing, and areas prohibited from shellfish harvest due to poor water quality) are depicted as an aquaculture restriction zone (ARZ) within the pond. The available Psammowassent soils would best support on-the-bottom production methods, while the Sulfiwassents would best support floating culture production. The Haplowassents could also support aquaculture development, however a portion of these soils have also been identified as prime areas for oyster restoration activities by habitat suitability index (HSI) modeling. The white polygons within the Psammowassent soils represent existing bivalve aquaculture lease areas………………………………………………………………134

xvii

MANUSCRIPT – I: SUBAQUEOUS SOILS AND COASTAL

ACIDIFICATION: A HYDROPEDOLOGY PERSPECTIVE WITH

IMPLICATIONS FOR CALCIFYING ORGANISMS

Published in the Soil Science Society of America Journal, March 13, 2015

Brett M. Still1, Mark H. Stolt1

1Department of Natural Resources Science, University of Rhode Island

1 Greenhouse Road, Kingston, RI 02881

ABSTRACT

In the coastal zone, biological and biogeochemical processes, influenced by anthropogenic inputs, drive pH variability and contribute to coastal acidification.

Spatial patterns of these processes across coastal estuaries are unknown. In this study, we used a hydropedological approach to assess the spatial variability of coastal acidification within two coastal lagoons and embayments in Rhode Island by measuring oyster shell dissolution, pH within the water column, and pore water pH within the upper 5 cm of the underlying subaqueous soils. Sampling and monitoring sites were stratified based on submerged soil-landscape types mapped at the Great

Group level as Haplowassents, Sulfiwassents, and Psammowassents. We found that pore water pH varied significantly among soils and with depth. Median pore water pH was significantly greater in sandy, low organic matter content Psammowassents (7.97) than the finer textured, higher soil organic matter content Sulfiwassents (7.35), and the

Haplowassents (6.57) that receive groundwater discharge from the surrounding subaerial soils. Juvenile calcifying organisms can experience acidic stress at pH values below 7.6; thus, current pH values within the upper few centimeters of Sulfiwassents and Haplowassents may be low enough to impact recently set juvenile calcifying organisms inhabiting these soils. Consequently, mean shell loss during a 4-wk period was significantly greater in the Sulfiwassents (1.54) than the Psammowassents

(0.96%), with the greatest shell loss (18.62%) in one of our Haplowassent sites with groundwater discharge. Our research suggests that measures of pore water pH and shell dissolution may be helpful in developing soil interpretations regarding the effects of coastal acidification on calcifying organisms.

INTRODUCTION

One of the more novel areas of hydropedological research during the last two decades has been studies of estuarine subaqueous soils (Stolt et al., 2007; Stolt and

Rabenhorst, 2011; Rabenhorst and Stolt, 2012). Although marine scientists had been studying substrates in the coastal zone for many decades, their studies were based on few if any spatial relationships. Early subaqueous soils research showed that shallow subtidal substrates could be inventoried and characterized using soil survey and pedological approaches (Demas et al., 1996; Bradley, 2001; Bradley and Stolt, 2003).

Later subaqueous soils research turned to developing use and management models using the subaqueous soil survey information for issues such as submerged aquatic vegetation restoration (Bradley and Stolt, 2006), C accounting (Jespersen and Osher,

2007; Payne, 2007), and shellfish aquaculture siting (Salisbury, 2010). As such, in their national Coastal and Marine Ecological Classification Standard (CMECS), the

Federal Geographic Data Committee (2012) recommended use of the subaqueous soils approach to classify shallow substrates for use and management interpretations:

“the Soil Geographic Data Standard, FGDC-STD-006 (FGDC 1997) and Keys to Soil Taxonomy (Soil Survey Staff, 2010) together provide more detailed classification options for classifying soils with many hundreds of descriptors that have been used in soil science for decades. Users should consider these sources and approaches when classifying substrate in these areas. It is recommended that a soils approach be used if a more detailed classification is needed for interpreting use and management of shallow water substrates.” (p. 102)

Because the CMECS is a Federal Geographic Data Committee (FGDC) approved standard, and the authors recommended FGDC-STD-006 for classifying shallow water substrates if use and management is an issue in shallow subtidal systems, any project

having US federal funding is expected to follow these recommendations. This suggests that the hydropedological approach to classify shallow subtidal substrates will be more widely used in the coming decades. One of the most pressing estuarine use and management issues with which subaqueous soils research may be able to assist now and in the near future is coastal acidification.

Anthropogenic activities have increased the atmospheric concentration of CO2 by

40% since the industrial revolution (Hartmann et al., 2013). As much as 30% of this

CO2 is being absorbed by the surface ocean (Doney, 2010; Rhein et al., 2013), leading to a decrease in both the surface ocean pH and the aragonite–calcite saturation state

(Feely et al., 2009). Continued CO2 absorption is projected to lower the open-ocean pH approximately 0.35 units by the end of the century. In coastal ecosystems, processes including watershed development, atmospheric deposition, freshwater nutrient inputs, upland soil erosion and weathering of geologic materials, excessive nutrient additions from agricultural and urban runoff, substrate and water column production, and respiration all interact at multiple spatial and temporal scales, resulting in order of magnitude greater pH variability within the coastal zone than the open ocean (Waldbusser et al., 2004; Waldbusser and Salisbury, 2014; Aufdenkampe et al., 2011; Cai et al., 2011; Hofmann et al., 2011; Duarte et al., 2013; Nixon et al.,

2015). Freshwater flows to coastal systems have also been shown to lower pH and aragonite saturation states by increasing the amount of poorly buffered surface or groundwater released into coastal systems, adding to the spatial complexity (Spiteri et al., 2006; Salisbury et al., 2008). As such, coastal zone ecosystems represent a complex mosaic of interactions between the upland landscape and coastal ocean.

These linkages dictate that the drivers associated with coastal acidification are much more complicated than the open ocean (Aufdenkampe et al., 2011; Cai et al., 2011;

Duarte et al., 2013; Nixon et al., 2015; Waldbusser and Salisbury, 2014) and that coastal acidification has the potential to have severe consequences for a range of calcifying marine organisms (Orr et al., 2005; Doney et al., 2009). Thus, understanding these complexities is critical to developing any long-term use and management plan focused on coastal management issues such as bivalve aquaculture development, shellfish restoration activities, and management of wild shellfish stocks.

Several CO2 manipulation mesocosm experiments have shown that commercially important bivalve species, including eastern oysters (Crassostrea virginica), hard clams (Mercenaria mercenaria), and bay scallops (Argopecten irradians) respond negatively to acidification (Ringwood and Keppler, 2002; Green et al., 2004, 2009;

Miller et al., 2009; Beniash et al., 2010; Talmage and Gobler, 2010; Waldbusser et al.,

2010, 2011). These responses include reduced growth, changes in shell structure, shell dissolution, poor larval development, and increased juvenile mortality. This collective body of studies indicates that early life stages are more susceptible due to greater surface area exposure to acidic conditions and that these effects can occur at pH values even as high as 7.6 depending on the aragonite saturation, salinity, and temperature.

Calcifying organisms that inhabit the coastal zone, like commercial bivalves, will probably be at risk before open-ocean organisms due to additional sources of acidification operating in the coastal zone that are not operating in the open ocean.

This accelerated acidification in the coastal zone has the potential to cause major

disruptions in ecosystem function and economic systems that rely on robust shellfish populations (Cooley and Doney, 2009).

Shallow, well-mixed coastal systems have a complex coupled water column and substrate biogeochemistry that drives fluxes in pH and bioavailable saturation states within the water column and shallow substrates (Fenchel and Riedl, 1970; Stahl et al.,

2006; Soetaert et al., 2007). The biogeochemical processes that drive these fluxes are different within different substrate types and result in complex spatial patterns of acidification within estuarine systems. A review and analysis of organic matter enrichment and substrate toxicity by Hargrave et al. (2008) demonstrated that hydrodynamics and local sedimentation and erosional dynamics determine the spatial patterns of organic enrichment, which can increase sulfide oxidation and in turn lower the pH, resulting in substrate conditions that are toxic to marine benthic infauna

(invertebrates residing within marine substrates).

These interactions make it difficult to project how coastal systems, as a whole, will respond to increasing acidification (Duarte et al., 2013). A review of existing long-term pH monitoring data from several estuaries conducted by Duarte et al. (2013) showed that no clear pattern emerges when assessing how coastal systems will respond to acidification stressors because the drivers of acidification in the coastal zone affect the pH variability to a greater extent than atmospheric CO2 alone. As such, research needs to be completed at scales finer than the ecosystem levels (103 m) described by Duarte et al. (2013) and more on the order of habitat scales (Guarinello et al., 2010; Shumchenia and King, 2010; Stolt et al., 2011; Oakley et al., 2012). Our previous research on subaqueous soils has shown that there are spatial relationships

between soil types and geographic areas within an estuary that allow mapping of the subaqueous soils using submerged soil landscape models (Bradley and Stolt, 2003;

Osher and Flannagan, 2007) that are at habitat scales (Stolt et al., 2011). By stratifying the shallow subtidal estuary into different soil types via a soil survey, we have been able to identify the best place for restoration of submerged aquatic vegetation, oyster aquaculture, and C sequestration and to identify areas that may be problematic if dredge materials are placed on the upland (Bradley and Stolt, 2006; Payne, 2007;

Pruett, 2010; Salisbury, 2010; Millar et al., 2015). The objective of this research was to begin to understand the spatial relationships between soils and coastal acidification within an estuary, to understand some of the driving forces that are resulting in acidification, and to identify soil habitats where coastal acidification may be the most problematic currently and in the immediate future.

METHODOLOGY

Study Sites

We conducted the sampling and field portions of these studies in Greenwich Bay,

Wickford Harbor, Ninigret Pond, and Quonochontaug Pond in Rhode Island (Fig. 1).

These embayments and coastal ponds represent a range of conditions and regulatory mandates associated with the management of coastal resources, including shellfish and bivalve aquaculture development. This includes a range of ground- and freshwater sources, proximity to urban impacts, tidal fluctuations, and eutrophication and water quality issues that could influence coastal acidification. The physical characteristics of these estuaries also differ significantly.

Greenwich Bay (1200 ha) is a sub-estuary located along the western shore of the larger Narragansett Bay estuary (Fig. 1). Greenwich Bay consists of ice-marginal alluvial and lacustrine fans on the western shore and submerged delta plain deposits to the north (Oakley and Boothroyd, 2006). This bay consists of the main bay section and five shallow coves (Warwick, Apponaug, Greenwich, Buttonwoods, and Brushneck coves). At mean low water, the average depth of the central basin is 2.7 m and 1 to 2 m in the coves (Erikson, 1998), with a tidal fluctuation between 0.9 and 1.2 m

(Kennedy and Lee, 2003).

Wickford Harbor is smaller and less urbanized than Greenwich Bay. Located on an outwash plain, the soils surrounding this bay are predominantly sandy acidic outwash materials. The largest freshwater input into Wickford Harbor is Mill Creek, which flows into Mill Cove in the northwest corner of the bay. The average water

depth in the bay at mean low tide is 1.5 m, and the tidal fluctuation is approximately 1 m.

Ninigret Pond (677 ha) and Quonochontaug Pond (312 ha) are both coastal lagoons located on the southern shore of Rhode Island (Fig. 1). We chose these ponds because they represent typical coastal lagoon estuarine systems where a partially enclosed barrier spit separates the lagoon from the open ocean. Both Ninigret and

Quonochontaug ponds have inlets that are permanently open, allowing an exchange of seawater twice a day with Block Island Sound to the south. The major geologic depositional environments within the ponds are flood tidal deltas, washover fans, and lagoon basins (Boothroyd et al., 1985; Bradley and Stolt, 2003). Both lagoons are classified as microtidal (<2-m tidal range), mixed energy, and wave dominated

(Davies, 1964; Hayes, 1979). The average salinities of Ninigret and Quonochontaug ponds are 28 and 31 g/kg, respectively (Boothroyd et al., 1985). Sediment input into both ponds is relatively low and occurs through tidal inlets, storm surge channels that cut through barrier spits, and overwash channels that transport sand over and through spits and into the lagoons. The tidal range of Ninigret Pond is 10 cm, while the tidal range of Quonochontaug Pond is 56 cm (Lee and Olsen, 1985). Ninigret Pond has an average depth of 1.2 m, while Quonochontaug Pond is deeper at 1.8 m (Boothroyd et al., 1985).

Soil Characterization

We focused our analyses of pore water pH and shell dissolution within the most dominant subaqueous soil landscape units mapped in Rhode Island. We sampled 13 sites representing both high-energy (depositional shoreline [DS], washover fan flat

[WFF], platform [P], and mainland beach [MB]) and low-energy (bay floor [BF], lagoon bottom [LB], and mainland cove [MC]) subaqueous soil landscapes across the two coastal ponds and two larger embayments (Fig. 1) (Table 1). At the Great Group level, the soils classify as Psammowassents (four sites), Sulfiwassents (seven sites), and Haplowassents (two sites) (Table 1). In general, Psammowassents are associated with high-energy subaqueous soil landscapes and are characterized as having sandy textures throughout and low accumulations of organic matter compared with the other soils we sampled. Sulfiwassents are associated with low-energy depositional subaqueous soil landscapes and are dominated by fine textures, high organic matter accumulation, and higher total sulfides than Psammowassents. Haplowassents are typically associated with mainland beach subaqueous soil landscapes and can have a broad range of characteristics and morphologies. The two Haplowassents we sampled for this project were sandy with low organic matter accumulation and receive groundwater discharge from the surrounding uplands (Table 1).

We collected a soil core at each location using a vibracore (7.6-cm aluminum tube vibrated into the soil; Lanesky et al., 1979) or Macaulay peat sampler, depending on the bottom type. Vibracores were sealed on both ends in the field, brought back to the laboratory, and kept at 4C until being opened for description and characterization.

Macaulay samples were described in the field, and samples were bagged and placed on ice to minimize sulfide oxidation, transported to the laboratory, and placed in the freezer at −16C. Soil characterization included the particle size distribution (Gee and

Bauder, 1986), bulk density (McVey et al., 2012), potential acidity (8-wk incubation

pH) (Soil Survey Staff, 2010), and organic matter and CaCO3 contents (Rabenhorst,

1988; Heiri et al., 2001). pH Sampling

Using a randomized sampling design, we conducted soil pH characterization using an ion sensitive field-effect transistor (ISFET) pH probe (0.01) (Spectrum

Technologies, IQ-150) within the near subtidal zone of the most dominant subaqueous soil landscapes throughout the ponds and embayments. Samples were collected during the summer larval settlement period to assess the variation of pH within these soils during a period when recently metamorphosed bivalve larvae shift from pelagic to benthic life stages as recently set juveniles. We used these pH data as an indicator of biogeochemical activity associated with potential drivers of coastal acidification.

We collected five replicate undisturbed soil cores within each sampling location using a modified plunge-corer fabricated from a 60-cm3 polycarbonate syringe (2.5 by

13 cm). The open-bottom corer was plunged into the soil surface to a depth of 10 cm, which effectively captured an 8-cm soil core and 2 cm of the overlying water. The corer was capped with a no. 6.5 rubber stopper predrilled with a 0.5-cm hole that was covered to create suction. The core was slowly drawn from the soil surface and rapidly capped on the bottom to prevent sample loss. Immediately following recovery, the

ISFET pH probe was inserted into the core, and we recorded the pH and temperature of the overlying water and at 1-cm increments within the core to a depth of at least 5 cm. Between each replicate sample, the ISFET pH probe was calibrated using standard

NBS buffers at ambient water temperature to ensure accurate readings and reduce probe drift. When reporting mean values for these pH data, we calculated the mean H+

concentration from the measured ISFET probe values and converted these mean concentrations to the logarithmic pH scale by calculating the negative logarithm of the mean H+ concentration.

Shell Dissolution

To assess oyster shell dissolution rates within the different mapped subaqueous soil landscape units, we conducted a shell dissolution experiment by deploying three replicate vinyl-coated polyester screen bags that contained 10 juvenile oyster shell valves. The shell bags were deployed in a subset of the dominant soil landscapes in

Greenwich Bay (BF, DS, and P) and Ninigret Pond (MBGW1, LB, and WFF). Each mesh bag measured 12.5 by 26 cm and was configured to hold one single oyster shell within each of 10 sewn pockets. The bags were anchored to the soil surface in an “L” configuration, so five shells were buried just below the soil surface and the other five shells were located in the water column just above the soil water interface.

Before deployment, we cleaned and prepped each individual oyster shell with a nylon bristle brush to remove biofouling and encrusting organisms. Once cleaned, we dried the shells to a constant weight (105C) and recorded the weight to the nearest 10 mg using a Mettler Toledo AB104 balance for each shell used in the experiment (180 total shells) ( X = 2.5416 g  0.076 SE). We randomly selected each shell, then labeled and placed one shell within each of the 10 pockets within each bag so each shell had a unique ID. The bags were deployed a total of 29 d from 2 through 30

August and were subsequently recovered from the field. Upon recovery, we carefully cleaned each shell to remove biofouling and dried the clean shells to a constant weight following the same protocols. The relative weight loss percentage was calculated for

each shell and used to determine the mean shell loss percentage during the deployment period.

Statistical Analysis

To characterize the variation of the pore water pH data, we conducted statistical analyses using SigmaPlot 11.0 (Systat Software). We used parametric and nonparametric Kruskal–Wallis ANOVA tests to compare the median pore water pH among the soils. We investigated the relationships between soil characteristics and pore water pH using Pearson product moment correlations. To explore the relationship between oyster shell loss and pore water pH across the different sampled soils, we performed a linear mixed effects model using the R statistical environment 2.14.1 (R

Core Team, 2011) and nlme (Pinheiro et al., 2014). As fixed effects, we included the mean pore water pH of the soil and water column where the oyster shells were located

(calculated as the mean H+ concentration converted to the pH scale), the initial shell weight, the treatment effect (soil vs. water column), and the interaction between the treatment effect and the mean pore water pH. Because the shells were contained within replicate bags at each site, we included the bag ID (n = 15) as a random effect in the model to account for the spatial correlation of the nested sampling design.

We followed the model selection and validation procedures outlined by Zuur et al. (2007) and Zuur et al. (2009) to first identify the optimal random structure and subsequently identify the optimal fixed structure. We used Akaike information criteria to identify the optimal random structure and likelihood ratio testing, using maximum likelihood estimation, to determine the optimal fixed structure for the model. Final model parameter estimates were obtained using restricted maximum likelihood

estimation. The final model was validated to check the underlying assumptions by plotting the residuals vs. fitted values to assess violations of homoscedasticity and investigating histograms and Q–Q plots to assess violations of normality (Zuur et al.,

2010).

RESULTS AND DISCUSSION

Pore Water and Water Column pH

The overlying water column pH from across all soil landscapes was variable and was higher than the pore water pH within the upper 5 cm of the soils that we sampled

(Table 2). When we pooled the sites based on Great Group classification, we observed statistically different pore water pH across the three Great Groups (Psammowassents,

Sulfiwassents, and Haplowassents) (H = 192.075, df2, P < 0.001) (Fig. 2; Table 3). We also observed statistically significant differences in the pore water pH across the sampled depths within each Great Group (Psammowassents, H = 29.435, df5, P <

0.001; Sulfiwassents, H = 90.428, df5, P < 0.001; Haplowassents, H = 39.168, df5, P <

0.001) (Table 4).

In the Psammowassents, the median pore water pH was 7.97 (0.02 SE) and the pore water pH decrease with depth was minimal, such that only the water column pH at the interface (0 cm) was significantly different than the soil pore water pH measured at lower depths (Dunn’s test, P < 0.05) (Table 4). Psammowassents occur in high- energy subaqueous landscape units such as flood-tidal deltas or washover fans. The sand-dominated particle size allows the surface water to enter and exchange with the soil pore water to a greater degree than in the finer textured soils such as

Sulfiwassents. Evidence of this increased exchange was observed in the subaqueous

soil temperature data collected by Salisbury and Stolt (2011) that showed, even at depths of 25 cm below the soil surface, that the water column affected the soil temperature more in sandy soils than in finer textured soils. In addition,

Psammowassents tend to have a higher density of infaunal macroinvertebrates that can further increase water exchange through burrowing and feeding activities depending on the benthic infaunal communities present within the soils (Rhoads and Boyer,

1982; Stahl et al., 2006; Hargrave et al., 2008). The water column–pore water exchange tends to keep the pore water pH values above 7.6 within the upper 5 cm

(Fig. 2). Pore water pH values below 7.6 in estuarine waters with high salinity

(polyhaline, 18– 30 g/kg) have been shown to induce physiological stress in calcifying organisms, including commercial bivalves, when accompanied by low aragonite saturation states (Green et al., 2004, 2009; Miller et al., 2009; Beniash et al., 2010;

Talmage and Gobler, 2010; Waldbusser et al., 2010, 2011).

In the Sulfiwassents, the pore water pH decrease with depth was greater than we observed in the Psammowassents. In these soils, the water column pH at the interface

(0 cm) was significantly different than the soil pore water pH measured at multiple lower depths, indicating that these soils exhibit less surface water exchange than the

Psammowassents (Dunn’s test, P < 0.05) (Table 4). Sulfiwassents occur in low-energy depositional basins and coves. These fine-textured soils often accumulate high levels of organic matter and sulfides, resulting in lower habitat quality and possible toxic conditions for benthic infauna (Hargrave et al., 2008). Due to limited pore water exchange, biogeochemical cycling in the presence of O2, including organic matter oxidation and microbial sulfide oxidation, within these soils can reduce the pore water

pH below that of sandy soils. The median pore water pH of 7.35 (0.03 SE) that we observed in the Sulfiwassents was significantly lower than that of the

Psammowassents (H = 192.075, df2, P < 0.001) and was lower than the pore water pH physiologic stress level of 7.6 (Table 3). These data indicate that the reduced water column–pore water exchange and dominant biogeochemistry within Sulfiwassents will probably limit habitat quality and may reduce bivalve larval recruitment in these soils.

Recent research has shown that metamorphosed hard clams and soft-shell clams (Mya arenaria) demonstrate preferential settlement and increased survival in substrates with favorable geochemistry and reduced acidity and tended to avoid or had greater mortality in substrates that would classify as Sulfiwassents (Clements and Hunt, 2014;

Green et al., 2013).

The Haplowassents we sampled had the lowest median pore water pH of the three

Great Groups we investigated (6.57  0.09 SE), and was significantly lower than the

Psammowassents and Sulfiwassents (H = 192.075, df2, P < 0.001) (Table 3). The pore water pH within these soils was significantly lower than that of the overlying water and decreased significantly across multiple depths to a greater degree than the

Sulfiwassents (Dunn’s test, P < 0.05) (Table 4). Both Haplowassent sites showed evidence of groundwater discharge in the adjacent subaerial environment in the form of broadleaved cattail (Typha latifolia L.) and red maple (Acer rubrum L.), two freshwater wetland plants that have moderate to low salt tolerance. The subaerial soils up gradient of these sites are formed in granitic glacial drift deposits that have measured pH values in the subsoil typically around 4.5 to 5.0 (Rector, 1981). The

groundwater associated with these subaerial soils is also very acidic and had a strong influence on the pore water pH that we observed at these sites.

Shell Dissolution

For the shell dissolution experiment, we used a subset of sites (six) in Greenwich

Bay (GB) and Ninigret Pond (NP) (Table 5). Three sites are classified as

Psammowassents (GB-DS, GB-P, and NP-WFF), two sites are classified as

Sulfiwassents (GB-BF and NP-LB), and one site is classified as a Haplowassent (NP-

MBGW1) (Table 5). Results from the juvenile oyster shell dissolution experiment were consistent with the overall pattern that we observed in the pooled pore water pH data, with significant differences in mean shell loss observed across the Great Groups we tested (H = 98.643, df2, P < 0.001) (Fig. 3; Table 6). The greatest mean shell loss in both the soil (24.18  2.49%) and water column (13.88  1.85%) treatments occurred at NP-MBGW1 (the Haplowassent), which also had the lowest mean pore water and water column pH (Table 2). The mean shell loss was significantly greater in the silty soils (Sulfiwassents) (1.64  0.09%) than the sandy soils (Psammowassents)

(1.06  0.07%) that we sampled (H = 98.643, df2, P < 0.001) (Table 6). These data clearly show that the mechanisms driving acidification of the pore water of the Great

Groups we tested can have a strong influence on the dissolution of oyster shell materials within these soils. The Haplowassent that receives groundwater discharge from acidic subaerial soils resulted in 10 times greater mean shell loss than soils at the other sites. The Sulfiwassents had 1.5 times greater mean shell loss than the

Psammowassents.

The linear mixed effects model analysis indicated that the pore water pH had a significant negative relationship with shell loss; therefore, as the mean pore water pH decreased or became more acidic, the mean shell loss increased (P = 0.019) (Table 7).

The model also depicted a significant treatment effect in explaining the mean shell loss, indicating that the shells located within the water column treatment had less loss than the shells within the soil treatment, as is evident in Fig. 3 (P = 0.003). The initial shell weight also has a significant negative relationship with the mean shell loss within the model, indicating that the smaller oyster shells had greater loss than shells with a larger initial weight at the start of the experiment (P = 0.022) (Table 7). The model also shows a significant interaction between mean pore water pH and the soil vs. water column treatment, indicating that this relationship was variable across the sites; however, most sites showed greater shell loss within the soil treatment than the water column treatment (P = 0.003). This optimal model showed no clear violations of model assumptions when investigating diagnostic plots.

Pore Water pH and Soil Characteristics

Analysis of A-horizon characteristics including potential acidity (8-wk incubation pH), bulk density, CaCO3 content, and organic matter content for the subset of soils that we included in the shell dissolution experiment shows significant correlations with the mean soil pore water pH (Table 8) for all measured characteristics. These data also indicate that the soil characteristics that we measured are also highly correlated, pointing to coupled biogeochemistry within these soils that contributes to shell dissolution.

Implications

Our research suggests that the pH associated with some surface subaqueous soils within shallow estuaries in Rhode Island may be low enough to induce physiological stress, shell dissolution, and mortality of recently set juvenile bivalves. Haplowassents that receive groundwater discharge from acidic subaerial soils and Sulfiwassents had soil pore water pH profiles within the upper 5 cm low enough to negatively affect some commercial bivalve species, including possible reduced recruitment, growth, and survival. It is clear from our data that different subaqueous soils within coastal lagoons and embayments can exert different levels of pH stress on calcifying organisms based on the dominant morphology and biogeochemistry operating within these soils.

These results lend strong support for the development of soil acidification interpretations based on subaqueous soil classifications as a way to stratify the shallow subtidal estuary into a range of expected pH conditions and identify areas within these systems that produce greater pH stress for calcifying organisms including commercially important bivalves. Such interpretive maps can be a valuable tool for spatial planning of aquaculture development, identifying shellfish restoration sites, management of wild stocks, siting spawner sanctuaries, or identifying areas that may be resilient to increased acidification of the estuary based on soil morphologies and geochemistry.

ACKNOWLEDGEMENTS

This project was funded by The Nature Conservancy’s Global Marine Program and the Rhode Island Agricultural Experiment Station through funds projected for

Multistate Project NE-1038. We would like to thank Dr. Jose Amador, Dr. Adam

Smith, Pamela Loring, Andrew Paolucci, Kris Plante, and Mason Garfield from the

URI Natural Resources Science Department, as well as Jim Turenne and Maggie

Payne from the USDA–NRCS for field and laboratory support during this project.

This paper is a contribution of the Rhode Island Agricultural Experiment Station (no.

5416).

CITATIONS

Aufdenkampe, A.K., E. Mayorga, P.A. Raymond, J.M. Melack, S.C. Doney, S.R. Alin, and K. Yoo. 2011. Riverine coupling of biogeochemical cycles between land, oceans, and atmosphere. Front. Ecol. Environ 9:53–60.

Beniash, E., A. Ivanina, N. Lieb, I. Kurochkin, and I. Sokolova. 2010. Elevated level of carbon dioxide affects metabolism and shell formation in oysters Crassostrea virginica (Gmelin). Mar. Ecol. Prog. Ser. 419:95–108.

Boothroyd, J.C., N.E. Friedrich, and S.R. McGinn. 1985. Geology of microtidal coastal lagoons: Rhode Island. Mar. Geol. 63:35–76.

Bradley, M.P. 2001. Subaqueous soils and subtidal wetlands in Rhode Island. M.S. thesis. Dep. of Natural Resources Science, Univ. of Rhode Island, Kingston.

Bradley, M.P., and M.H. Stolt. 2003. Subaqueous soil–landscape relationships in a Rhode Island estuary. Soil Sci. Soc. Am. J. 67:1487–1495.

Bradley, M.P., and M.H. Stolt. 2006. Landscape-level seagrass–sediment relations in a coastal lagoon. Aquat. Bot. 84:121–128.

Cai, W.J., X. Hu, W.J. Huang, M.C. Murrell, J.C. Lehrter, S.E. Lohrenz, and G.C. Gong. 2011. Acidification of subsurface coastal waters enhanced by eutrophication. Nat. Geosci. 4:766–770.

Clements, J.C., and H.L. Hunt. 2014. Influence of sediment acidification and water flow on sediment acceptance and dispersal of juvenile soft-shell clams (Mya arenaria L.). J. Exp. Mar. Biol. Ecol. 453:62–69.

Cooley, S.R., and S.C. Doney. 2009. Anticipating ocean acidification’s economic consequences for commercial fisheries. Environ. Res. Lett. 4:024007.

Davies, J.L. 1964. A morphogenetic approach to world shorelines. Z. Geomorphol. 8:127–142.

Demas, G.P., M.C. Rabenhorst, and J.C. Stevenson. 1996. Subaqueous soils: A pedological approach to the study of shallow-water habitats. Estuaries 19:229–237.

Doney, S.C. 2010. The growing human footprint on coastal and open-ocean biogeochemistry. Science 328:1512–1516.

Doney, S.C., V.J. Fabry, R.A. Feely, and J.A. Kleypas. 2009. Ocean acidification: The other CO2 problem. Annu. Rev. Mar. Sci. 1:169–192.

Duarte, C.M., I.E. Hendriks, T.S. Moore, Y.S. Olsen, A. Steckbauer, L. Ramajo, and M. McCulloch. 2013. Is ocean acidification an open-ocean syndrome? Understanding anthropogenic impacts on seawater pH. Estuaries Coasts 36:221– 236.

Erikson, L. 1998. Flushing times of Greenwich Bay, RI: Estimates from freshwater inputs. M.S. thesis. Dep. of Ocean Engineering, Univ. of Rhode Island, Narragansett.

Federal Geographic Data Committee. 1997. FGDC-STD-006: Soil geographic data standard. FGDC, Reston, VA.

Federal Geographic Data Committee. 2012. FGDC-STD-018-2012: Coastal and marine ecological classification standard. FGDC, Reston, VA. https://www.fgdc.gov/standards/projects/FGDC-standards-projects/cmecs- folder/CMECS_Version_06-2012_FINAL.pdf.

Feely, R.A., S.C. Doney, and S.R. Cooley. 2009. Ocean acidification: Present conditions and future changes in a high-CO2 world. Oceanography 22:36–47.

Fenchel, T.M., and R.J. Riedl. 1970. The sulfide system: A new biotic community underneath the oxidized layer of marine sand bottoms. Mar. Biol. 7:255–268.

Gee, G.W., and J.W. Bauder. 1986. Particle-size analysis. In: A. Klute, editor, Methods of soil analysis. Part 1. 2nd ed. SSSA Book Ser. 5. ASA and SSSA, Madison, WI. p. 383–411. doi:10.2136/sssabookser5.1.2ed.c15

Green, M.A., M.E. Jones, C.L. Boudreau, R.L. Moore, and B.A. Westman. 2004. Dissolution mortality of juvenile bivalves in coastal marine deposits. Limnol. Oceanogr. 49:727–734.

Green, M.A., G.G. Waldbusser, L. Hubazc, E. Cathcart, and J. Hall. 2013. Carbonate mineral saturation state as the recruitment cue for settling bivalves in marine muds. Estuaries Coasts 36:18–27.

Green, M.A., G.G. Waldbusser, S.L. Reilly, K. Emerson, and S. O’Donnell. 2009. Death by dissolution: Sediment saturation state as a mortality factor for juvenile bivalves. Limnol. Oceanogr. 54:1037–1047.

Guarinello, M.L., E.J. Shumchenia, and J.W. King. 2010. Marine habitat classification for ecosystem-based management: A proposed hierarchical framework. Environ. Manage. 45:793–806.

Hargrave, B.T., M. Holmer, and C.P. Newcombe. 2008. Towards a classification of organic enrichment in marine sediments based on biogeochemical indicators. Mar. Pollut. Bull. 56:810–824.

Hartmann, D.L., A.M.G. Klein Tank, M. Rusticucci, L.V. Alexander, S. Brönnimann, Y. Charabi, et al. 2013. Observations: Atmosphere and surface In: T.F. Stocker et al., editors, Climate change 2013: The physical science basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge Univ. Press, Cambridge, UK.

Hayes, M. O. 1979. Barrier island morphology as a function of tidal and wave regime. In: S.P. Leatherman, editor, Barrier islands from the Gulf of St. Lawrence to the Gulf of Mexico. Academic Press, New York. p. 1–27.

Heiri, O., A.F. Lotter, and G. Lemcke. 2001. Loss on ignition as a method for estimating organic and carbonate content in sediments: Reproducibility and comparability of results. J. Paleolimnol. 25:101–110.

Hofmann, G.E., J.E. Smith, K.S. Johnson, U. Send, L.A. Levin, F. Micheli, et al. 2011. High frequency dynamics of ocean pH: A multi ecosystem comparison. PLoS One 6(12):e28983.

Jespersen, J.L., and L.J. Osher. 2007. Carbon storage in the soils of a mesotidal Gulf of Maine estuary. Soil Sci. Soc. Am. J. 71:372–379.

Kennedy, S., and V. Lee. 2003. Greenwich Bay: An ecological history. Rhode Island Sea Grant Commun. Office, Univ. of Rhode Island, Narragansett.

Lanesky, D.E., B.W. Logan, R.G. Brown, and A.C. Hine. 1979. A new approach to portable vibracoring underwater and on land. J. Sediment. Petrol. 49:654–657.

Lee, V., and S. Olsen. 1985. Eutrophication and management initiatives for the control of nutrient inputs to Rhode Island coastal lagoons. Estuaries 8:191–202.

McVey, S., P.J. Schoeneberger, J. Turenne, M. Payne, D.A. Wysocki, and M.H. Stolt. 2012. Subaqueous soils (SAS) description. In: P.J. Schoeneberger et al., editors, Field book for describing and sampling soils. Version 3.0. Natl. Soil Surv. Ctr., Lincoln, NE. p. 2-97–2-119.

Millar, C.M., M.H. Stolt, A.A.O. Aduomih, and B. Still. 2015. Estuarine subaqueous soil organic carbon accounting: Sequestration and storage. Soil Sci. Soc. Am. J. 79:389–397.

Miller, A.W., A.C. Reynolds, C. Sorbino, and G.F. Riedel. 2009. Shellfish face uncertain future in high CO2 world: Influence of acidification on oyster larvae calcification and growth in estuaries. PLoS one 4:5.e5661.

Nixon, S.W., A.J. Oczkowski, M.E. Pilson, L. Fields, C.A. Oviatt, and C.W. Hunt. 2015. On the response of pH to inorganic nutrient enrichment in well-mixed coastal marine waters. Estuaries Coasts (in press).

Oakley, B.A., J.D. Alvarez, and J.C. Boothroyd. 2012. Benthic geologic habitats of shallow estuarine environments: Greenwich Bay and Wickford Harbor, Narragansett Bay, Rhode Island, USA. J. Coast. Res. 28:760–773.

Oakley, B.A., and J. Boothroyd. 2006. Benthic geologic habitats of Greenwich Bay, Rhode Island. Poster presented at the 41st Annual Meeting of the Northeastern Section of the Geologic Society of America, Harrisburg, PA. 20–22 Mar. 2006.

Orr, J.C., V.J. Fabry, O. Aumont, L. Bopp, S.C. Doney, R.A. Feely, et al. 2005. Anthropogenic ocean acidification over the twenty-first century and its impact on calcifying organisms. Nature 437:681–686.

Osher, L.J., and C.T. Flannagan. 2007. Soil/landscape relationships in a mesotidal Maine estuary. Soil Sci. Soc. Am. J. 71:1323–1334.

Payne, M.K. 2007. Landscape-level assessment of subaqueous soil and water quality in shallow embayments in southern New England. M.S. thesis. Univ. of Rhode Island, Kingston.

Pinheiro, J., D. Bates, S. DebRoy, D. Sarkar, and R Core Team. 2014. nlme: Linear and nonlinear mixed effects models. R package version 3.1-117. Inst. Stat. Math., Wirtschaftsuniversität Wien, Vienna. http://CRAN.R-project.org/package=nlme.

Pruett, C.M. 2010. Interpretations of estuarine subaqueous soils: Eelgrass restoration, carbon accounting, and heavy metal accumulation. M.S. thesis. Univ. of Rhode Island, Kingston.

R Core Team. 2011. R: A language and environment for statistical computing. R Found. Stat. Comput., Vienna.

Rabenhorst, M.C. 1988. Determination of organic and carbonate carbon in calcareous soils using dry combustion. Soil Sci. Soc. Am. J. 52:965–969.

Rabenhorst, M.C., and M.H. Stolt. 2012. Subaqueous soils: Pedogenesis, mapping, and applications. In: H. Lin, editor, Hydropedology: Synergistic integration of soil science and hydrology. Academic Press, San Diego. p. 173–204.

Rector, D.D. 1981. Soil survey of Rhode Island. US Gov. Print. Office, Washington, DC.

Rhein, M., S.R. Rintoul, S. Aoki, E. Campos, D. Chambers, R.A. Feely, et al. 2013. Observations: Ocean. In: T.F. Stocker et al., editors, Climate change 2013: The physical science basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge Univ. Press, Cambridge, UK.

Rhoads, D.C., and L.F. Boyer. 1982. The effects of marine benthos on physical properties of sediments: A successional perspective. Top. Geobiol. 100:3–52.

Ringwood, A.H., and C.J. Keppler. 2002. Water quality variation and clam growth: Is pH really a non-issue in estuaries? Estuaries 25:901–907.

Salisbury, A.R. 2010. Developing subaqueous soil interpretations for Rhode Island estuaries. M.S. thesis. Univ. of Rhode Island, Kingston.

Salisbury, A., and M.H. Stolt. 2011. Estuarine subaqueous soil temperature. Soil Sci. Soc. Am. J. 75:1584–1587.

Salisbury, J., M.A. Green, C. Hunt, and J. Campbell. 2008. Coastal acidification by rivers: A threat to shellfish? Eos Trans. AGU 89:513–514.

Shumchenia, E.J., and J. King. 2010. Evaluation of sediment profile imagery as a tool for assessing water quality in Greenwich Bay, Rhode Island, USA. Ecol. Indic. 10:818–825.

Soetaert, K., A.F. Hofmann, J.J. Middelburg, F.J.R. Meysman, and J. Greenwood. 2007. The effect of biogeochemical processes on pH. Mar. Chem. 105:30–51.

Soil Survey Staff. 2010. Keys to Soil Taxonomy. 11th ed. US Gov. Print. Office, Washington, DC.

Spiteri, C., P. Regnier, C.P. Slomp, and M.A. Charette. 2006. pH-dependent iron oxide precipitation in a subterranean estuary. J. Geochem. Explor. 88:399–403.

Stahl, H., A. Glud, C.R. Schroder, I. Klimant, A. Tengberg, and R.N. Glud. 2006. Time-resolved pH imaging in marine sediments with a luminescent planar optode. Limnol. Oceanogr. Methods 4:336–345.

Stolt, M., M. Bradley, J. Turenne, M. Payne, and E. Scherer. 2007. From land to shallow water: The evolution of soil mapping in the MapCoast Partnership. 41N 4(1):4.

Stolt, M., M. Bradley, J. Turenne, M. Payne, E. Scherer, G. Cicchetti, et al. 2011. Mapping shallow coastal ecosystems: A case study of a Rhode Island lagoon. J. Coast. Res. 27(6A):1–15.

Stolt, M.H., and M.C. Rabenhorst. 2011. Subaqueous soils. In: P.M. Huang et al., editors, Handbook of soil science. 2nd ed. CRC Press, Boca Raton, FL. p. 36-1– 36-14.

Talmage, S.C., and C.J. Gobler. 2010. Effects of past, present, and future ocean carbon dioxide concentrations on the growth and survival of larval shellfish. Proc. Natl. Acad. Sci. 107:17246–17251.

Waldbusser, G., H. Bergschneider, and M.A. Green. 2010. Size-dependent pH effect on calcification in post-larval hard clam Mercenaria spp. Mar. Ecol. Prog. Ser. 417:171–182.

Waldbusser, G.G., R.L. Marinelli, R.B. Whitlatch, and P.T. Visscher. 2004. The effects of infaunal biodiversity on biogeochemistry of coastal marine sediments. Limnol. Oceanogr. 49:1482–1492.

Waldbusser, G.G., and J.E. Salisbury. 2014. Ocean acidification in the coastal zone from an organism’s perspective: Multiple system parameters, frequency domains, and habitats. Annu. Rev. Mar. Sci. 6:221–247.

Waldbusser, G.G., E.P. Voigt, H. Bergschneider, M.A. Green, and R.I.E. Newell. 2011. Long-term trends in Chesapeake Bay pH and effects on biocalcification in the eastern oyster Crassostrea virginica. Estuaries Coasts 34:221–231.

Zuur, A.F., E.N. Ieno, and C.S. Elphick. 2010. A protocol for data exploration to avoid common statistical problems. Methods Ecol. Evol. 1:3–14.

Zuur, A.F., E.N. Ieno, and G.M. Smith. 2007. Analyzing ecological data. Springer, New York.

Zuur, A.F., E.N. Ieno, N. Walker, A.A. Saveliev, and G.M. Smith. 2009. Mixed effects models and extensions in ecology with R. Springer, New York.

TABLES

Table 1.1. Taxonomic classification and general characteristics for the surface horizon of the subaqueous soils we sampled for pore water pH. Data obtained from NRCS SSURGO database. Particle size percent based on USDA Textural Soil Classification. Subgroup Organic Series Site† Soil-landscape unit Clay Sand classification matter % GB-BF bay floor WH-BF Fluventic Sulfiwassents Pishagqua NP-LB 17 18 14 lagoon bottom QP-LB NP-MB mainland beach GB-DS depositional shoreline Fluventic Psammowassents Rhodesfolly 1 98 1.5 GB-P platform Thapto-histic Sulfiwassents Billington NP-MC mainland cove 17 18 14 Fluventic Psammowassents Massapog WH-DS depositional shoreline 0 97 1 NP-WFF

28 Sulfic Psammowassents Nagunt washover fan flat 1 98 1 QP-WFF NP-MBGW1‡ Aeric Haplowassents Napatree mainland beach 0 99 1 NP-MBGW2‡ † Sites are located in Greenwich Bay (GB), Wickford Harbor (WH), Ninigret Pond (NP), and Quonochontaug Pond (QP) in bay floor (BF), lagoon bottom (LB), depositional shoreline (DS), platform (P), mainland cove (MC), washover fan flat (WFF), and mainland beach (MB) soil landscape units ‡Sites with groundwater intrusion (NP-MBGW1 and 2) were sampled for pore water pH but not included in mixed model analysis

Table 1.2. Mean pore water pH pooled across depth (1–5 cm) and mean water column pH collected from subaqueous soil landscapes in Rhode Island. SE is the standard error of the mean. N represents the number of samples taken at each site. Mean pore water pH of the soil and water column was calculated as the mean hydrogen ion concentration converted to pH scale (–log of the mean H+ concentration). Pore water Water column Site† pH n pH n Greenwich Bay GB-BF 7.17 (0.02)‡ 50 7.65 (0.04) 10 GB-DS 7.92 (0.03) 25 8.05 (0.05) 5 GB-P 7.64 (0.05) 50 7.90 (0.04) 10 Ninigret Pond NP-LB 7.13 (0.03) 25 8.11 (0.04) 5 NP-MC 7.48 (0.03) 20 8.37 (0.03) 4 NP-MB 7.30 (0.05) 25 8.36 (0.08) 5 NP-MBGW1 5.67 (0.07) 50 7.18 (0.12) 10 NP-MBGW2 6.96 (0.08) 25 8.19 (0.02) 5 NP-WFF 7.91 (0.04) 50 8.00 (0.12) 10 Quonochontaug Pond QP-LB 7.02 (0.04) 25 8.27 (0.06) 5 QP-WFF 7.76 (0.05) 25 8.00 (0.12) 5 Wickford Harbor WH-BF 6.99 (0.03) 25 7.94 (0.14) 5 WH-DS 7.67 (0.07) 25 8.26 (0.04) 5 † Sites occur on bay floor (BF), lagoon bottom (LB), depositional shoreline (DS), platform (P), mainland cove (MC), washover fan flat (WFF), and mainland beach (MB) soil-landscape units. ‡ Standard error in parentheses.

Table 1.3. Kruskal–Wallis ANOVA on ranks (H, α = 0.05) for pore water pH samples pooled by Great Group. Great Group N pH Test statistic Psammowassents 149 7.97 (0.02) a† Sulfiwassents 264 7.35 (0.03) b H = 192.075, df2, P < 0.001 Haplowassents 90 6.57 (0.09) c † Median with standard error in parentheses. Medians followed by different letters are significantly different (Dunn’s test for multiple comparisons).

Table 1.4. Kruskal–Wallis ANOVA on ranks (H, α = 0.05) for pore water pH samples across depth for sites pooled by great group classification. Letters depict significant comparisons across depth within great groups (Dunn’s test for multiple comparisons). Depth N pH Test statistic cm Psammowassents 0 25 8.17 (0.03) a† 1 25 8.00 (0.04) ab 2 25 7.96 (0.05) ab H = 29.435, df5, P < 0.001 3 25 7.91 (0.05) b 4 25 7.87 (0.05) b 5 24 7.89 (0.05) b Sulfiwassents 0 44 8.06 (0.05) a 1 44 7.47 (0.05) bc 2 44 7.27 (0.05) bc H = 90.428 df5, P < 0.001 3 44 7.18 (0.05) bc 4 44 7.12 (0.06) c 5 44 7.17 (0.06) c Haplowassents 0 15 8.10 (0.15) a 1 15 6.65 (0.13) a 2 15 6.34 (0.15) bc H = 39.168 df5, P < 0.001 3 15 6.11 (0.18) bc 4 15 5.84 (0.18) bc 5 15 5.64 (0.20) c † Median with standard error in parentheses. Medians followed by different letters are significantly different (Dunn’s test for multiple comparisons).

Table 1.5. Subset of sites associated with the oyster shell dissolution experiment. Data represent surface horizon characteristics taken by vibracore or Macaulay samples and mean pH for the water column (WC) and upper 3 cm of the soil pore water (PW). Mean pore water pH of the soil and water column was calculated as the mean H+ concentration converted to the pH scale (−log of the mean H+ concentration). Bulk 8-wk WC PW Site† Classification Sand OM‡ CaCO3 density incubation pH pH pH % g/cm3 Greenwich Bay GB-BF Fluventic Sulfiwassents 38.77 8.83 4.18 0.40 4.88 7.65 7.21 GB-DS Fluventic Psammowassents 97.77 0.99 1.52 1.23 8.66 8.05 7.95 GB-P Fluventic Psammowassents 98.70 0.52 0.33 1.17 8.62 7.90 7.59 Ninigret Pond NP-MBGW1 Aeric Haplowassent 92.41 0.46 0.22 nd§ nd 7.18 6.01 NP-LB Fluventic Sulfiwassents 7.82 13.09 6.14 0.19 5.78 8.11 7.23 NP-WFF Sulfic Psammowassents 98.82 0.36 0.20 1.28 7.59 8.13 7.99

32 † Sites occur on bay floor (BF), lagoon bottom (LB), depositional shoreline (DS), platform (P), washover fan flat (WFF), and mainland beach

groundwater-influenced (MBGW) soil-landscape units. ‡ Organic matter by loss-on-ignition. § nd, not determined.

Table 1.6. Kruskal–Wallis ANOVA on ranks (H, α = 0.05) for mean shell loss percentage pooled by Great Group. Great Group N Shell loss Test statistic % Psammowassents 90 0.964 (0.05) a† H = 98.643, df2, P < 0.001 Sulfiwassents 59 1.542 (0.064) b Haplowassents 28 18.627 (1.80) c ‡ Median with standard error in parentheses. Medians followed by different letters are significantly different (Dunn’s test for multiple comparisons).

Table 1.7. Optimal linear mixed effects model results for fixed effects and random intercept terms included in the model. Mean pore water pH was calculated as the mean H+ concentration converted to the pH scale (−log of the mean H+ concentration). Variables Value SE df t value P value Fixed effects Intercept 4.86 1.40 129 3.47 0.000 Mean pore water pH −0.43 0.18 129 −2.38 0.018 Treatment (water column) 8.99 3.01 129 2.99 0.003 Initial shell weight −0.10 0.04 129 −2.32 0.022 Pore water pH  treatment −1.12 0.38 129 −2.99 0.003 Random effect (intercept) Bag ID SD = 0.28 Residual error = 0.40

Table 1.8. Correlation matrix for soil characteristics and mean pore water pH across sampled sites included in the shell loss experiment. Each comparison within the matrix provides the Pearson product moment correlation coefficient (r) and P value (n = 5). Mean pore water pH was calculated as the mean H+ concentration converted to the pH scale (−log of the mean H+ concentration). Organic Parameter Incubation pH† Bulk density CaCO 3 matter r P r P r P r P Mean pore water pH 0.97 0.007 0.96 0.009 −0.95 0.015 −0.96 0.011 Incubation pH 0.89 0.040 −0.85 0.069 −0.86 0.060 Bulk Density −0.98 0.005 −0.99 0.001 CaCO3 0.98 0.002 † Incubation pH data were collected during an 8-wk period.

FIGURES

MANUSCRIPT – 2: OYSTER GROWTH AND SURVIVAL ACROSS

SUBAQUEOUS SOIL LANDSCAPES AND ESTUARIES IN RHODE ISLAND

Manuscript will be sent to Journal of Shellfish Research for publication

Brett M. Still1, Alex R. Salisbury1, Mark H. Stolt1

1Department of Natural Resources Science, University of Rhode Island

1 Greenhouse Road, Kingston, RI 02881

ABSTRACT

Anthropogenic stressors have depleted oyster populations throughout many estuaries in the US and around the world. As a result, wild oyster populations cannot support the increased demand for oysters as a food resource without further affecting wild stocks. Aquaculture and restoration of existing stocks are seen as a possible solution to provide increased access to high value protein that cannot be supported by harvest of existing wild stocks. However, sustainable management of the growing aquaculture industry is needed to address use of limited space and stakeholder conflict within the coastal zone. In this study we used subaqueous soil classification and mapping as a tool for oyster aquaculture and restoration site selection. We conducted growth trials with juvenile eastern oysters (Crassostrea virginica) in dominant subaqueous soil types over five growing seasons within Rhode Island coastal estuaries. Using a linear mixed modeling statistical approach, we found that oysters grown in sandy firm substrates showed increased growth rates and survival when compared to oysters grown in silty substrates with low bearing capacity. These results suggest that substrate type may assist in identifying portions within estuaries that exhibit greater seston flux that has been shown to positively influence growth rate. These data suggest subaqueous soil maps are a good way to stratify estuarine substrates to identify preferred sites for oyster aquaculture development and restoration site selection.

INTRODUCTION

Commercially harvested bivalve populations around the world are experiencing multiple stressors associated coastal zone development, and increased harvest pressure due to the growing demand for the high valve protein that shellfish provide (FAO,

2012). The long history of destructive harvest practices and other anthropogenic stressors including deteriorating water quality, coastal eutrophication, sedimentation, disease, loss of habitat, and ocean/coastal acidification have led to dramatic declines in wild populations, and complete loss of ecosystem function within many estuaries worldwide (Alleway and Connell, 2015; Beck et al., 2011; Burge et. al., 2014;

Mackenzie, 2007; Mitchell et al., 2015; Rothschild et al., 1994; Seitz et al., 2014; Zu-

Ermgassen et al., 2013).

To overcome these stressors, communities are beginning to implement comprehensive ecosystem based management programs that seek to address management of wild stocks, aquaculture development, restoration, and user conflicts using best available science and comprehensive stakeholder involvement (Bricker et al., 2016; Mercer et al., 2015). Members of the scientific community are also increasingly supportive of bivalve aquaculture development to reduce harvest pressure on wild stocks, restore some lost ecosystem functions from depleted shellfish populations and provide improved economic conditions and access to high quality local food (Shumway et al., 2003). Identifying sites that can support these activities while providing optimal conditions for growth and recruitment is difficult without a systematic inventory of subtidal resources.

Currently there are few standardized classification systems that can be utilized to inventory these shallow subtidal areas (Stolt et al., 2011). One inventory approach is to use subaqueous soil mapping technologies and classification standards that have been developed and tested over the last 20 years (Bradley & Stolt, 2002; Bradley and

Stolt, 2003; Demas, 1993, Demas, 1998; Demas and Rabenhorst, 1999). Adoption of this approach by the USDA-NRCS has led to an expansion of the National

Cooperative Soil Survey soil mapping into shallow coastal waters of the United States.

As such, we now have the ability to systematically characterize these shallow subtidal landscapes, and provide a system of classification that can be used for the development of soil interpretations that can guide management decisions (Soil Survey

Staff, 2009; Soil Survey Staff, 2010).

Further support was provided for the subaqueous soils approach in the new national Coastal and Marine Ecological Classification Standards (CMECS). For shallow water, where detailed management and interpretations are required, CMECS

(FDGC, 2012) recommends that a subaqueous soil survey and classification approach be used for classifying substrates. These recommendations were based on the results from a suite of studies that showed that a subaqueous soil survey can be used for a range of coastal management purposes. Examples of ecosystem applications of subaqueous soil surveys include; eelgrass distribution (Bradley and Stolt, 2006), carbon accounting in shallow subtidal ecosystems (Jespersen and Osher, 2007; Pruett,

2010; Millar et al., 2015), estuarine water quality assessment (Payne, 2007), site selection for oyster and hard-shell clam aquaculture (Salisbury, 2010), and identification of areas more susceptible to coastal acidification (Still and Stolt, 2015).

Oysters are typically associated with sand, hard bottom and shell reef substrates and are absent most often from areas of fine substrate (silty and silty clay muds) and areas with high sedimentation rates (Brooks, 1996; Burrell, 1986; Sellers and Stanley,

1986; Shumway, 1996). Coarse textured substrates may not only provide a stable habitat, but also represent areas of greater current flow within estuaries, providing greater seston flux (food availability) that supports increased growth relative to fine textured substrates (Grizzle and Lutz, 1989; Rice, 1992; Rice and Pechenik, 1992).

Based on preliminary studies in RI, Salisbury (2010) suggested that subaqueous soil type may act as a surrogate for seston flux, and further explain variable growth rates and survival among different soil types. For example, lease areas located on coarse textured soils (Psammowassents) had higher growth rates than lease areas located on fine textured soils (Sulfiwassents), resulting in a greater proportion of oysters reaching market size in a shorter period of time (Salisbury, 2010). These results suggested that oyster aquaculturists should consider growth characteristics when considering a lease site location. However, these performance and growth data are not readily available and can be variable from year to year.

In this study, we build on previous work and further test soil-oyster aquaculture growth relationships over several growing seasons within the most dominant subaqueous soil landscapes within a range of Rhode Island estuaries. The objective was to determine if subaqueous soil maps can be utilized as a planning tool for aquaculture siting and restoration site selection. We conducted oyster growth research within Greenwich Bay, Wickford Harbor, Ninigret Pond, and Quonochontaug Pond in

Rhode Island, USA (Fig.1). These embayments and coastal lagoons represent a range

of physical characteristics and environmental conditions to test our hypothesis that certain soils are better suited for oyster growth, survival, and time to market size.

METHODOLOGY

Study sites Greenwich Bay (1,200 ha), is a sub-estuary, located along the western shore of

Narragansett Bay (Fig. 2.1). This bay consists of the main bay, and five shallow coves with estuarine substrate deposited over ice-marginal alluvial and lacustrine deposits on the western shore and submerged delta plain deposits to the north (Oakley and

Boothroyd, 2006). At mean low water, the average depth of the central basin is 2.7 m and 1-2 m in the coves (Erikson, 1998) with a tidal fluctuation between 0.9-1.2 m

(Kennedy and Lee, 2003). Currently there are no shellfish aquaculture leases within

Greenwich Bay, however the bay supports a viable commercial quahog fishery.

Wickford Harbor located approximately 10 km south of Greenwich Bay is smaller and less urbanized (Fig. 2.1). Soils surrounding Wickford Harbor are predominantly sandy acidic outwash materials. The largest freshwater input into

Wickford Harbor is Mill Creek, which flows into Mill Cove in the northwest corner of the bay. The average water depth at mean low tide is 1.5 meters, and the tidal fluctuation is approximately one meter. There are currently no shellfish aquaculture leases within the harbor, however a few shellfish aquaculture farms are located just northeast of the harbor entrance within Narragansett Bay.

Ninigret Pond (677 ha), and Quonochontaug Pond (312 ha) are both coastal lagoons located on the southern shore of Rhode Island (Fig. 2.1). We chose these ponds because they represent typical coastal lagoon estuarine systems where a partially enclosed barrier beach separates the lagoons from the open ocean. Both

Ninigret and Quonochontaug have inlets that are permanently open for navigation, allowing semidiurnal tidal exchange of seawater with Block Island Sound to the south.

The major geologic depositional environments within the ponds are flood tidal deltas, washover fans, and lagoon basins (Boothroyd et al., 1985; Bradley and Stolt, 2003).

Both lagoons are classified as microtidal (<2 m tidal range), mixed-energy, and wave dominated (Davies, 1964; Hayes, 1979). Sediment input into both ponds is relatively low and occurs through the permanent tidal inlets, and overwash channels that transport sand over and through the barrier beach into the lagoons during storm events.

The tidal range of Ninigret Pond is 10 cm while the tidal range of Quonochontaug

Pond is 56 cm (Lee and Olsen, 1985). Ninigret Pond has an average depth of 1.2 m while Quonochontaug Pond is deeper at 1.8 m (Boothroyd et al., 1985). A majority of both ponds are approved for shellfish harvest throughout the year, and support shellfish aquaculture operations (RIDEM, 2015).

Oyster Growth Trials

We investigated relationships between subaqueous soil landscapes and oyster growth within each pond and embayment. This work includes off-the-bottom data from previous work conducted in 2008 and 2009 by the URI Laboratory of Pedology and Soil Environmental Science (Salisbury, 2010). This collection of studies used the same methods and some of the same research sites in Ninigret and Quonochontaug ponds. We established twelve research sites (three within each waterbody) within the near-subtidal zone ranging in depth from 1.5 – 2 m (relative to MLW) to compare oyster growth across dominant subaqueous soil landscapes within each system. At each site in the off-the-bottom studies, we deployed three replicate standard aquatrays

(910 mm x 930 mm UV protected rigid polypropylene mesh tray with lid, supported by 38 mm dia. PVC supports) which were elevated approximately 20 cm above the

soil surface. At the start of each two-year growth trial, the aquatrays were planted with juvenile oysters ( = height 25 - 36 mm) at a density of approximately 250 oysters per square meter. In late fall, we measured total valve height from a randomly selected sample of 30 oysters from each aquatray to determine a mean growth rate within each tray (mm/day, n = 30) [(L2-L1)/(t2-t1)] (Abbe et al., 2003; Grizzle and Morin, 1989). In addition to growth rate, we calculated percent survival (number of live oysters out of total planted), and percent market size (number of oysters ≥ 76 mm out of the total sample size) at the end of each growing season within each site. To reduce excessive biofouling during the growing season, we cleaned the aquatrays with hand-held nylon brushes as needed.

In addition to the aquatray plots, we established direct on-the-bottom plots. Our initial plantings of juveniles in 2011 failed because of significant mortality. We re- established these on bottom plots in 2013 using 2nd year seed planted in 2012 to compare the difference in growth between oysters in aquatrays and oysters grown directly on the soil surface. We placed the oyster seed within UV stabilized polyethylene oyster grow-out bags to reduce mortality.

We intended to use the same seed source throughout the growth trials, however, given limited seed availability and issues with disease certificates required for permitting, we were forced to acquire seed from three different sources over the course of the growth experiments. In 2008 certified oyster seed was acquired from a local aquaculturalist in RI. At the start of the 2011 and 2012 growth trials, we acquired certified seed from two different hatcheries on Long Island, NY.

Water Quality

During the oyster growth experiments, we measured water quality parameters including dissolved oxygen, salinity, temperature, and pH with a multi-probe YSI

(YSI 556 MPS) within the water column (consistent with aquatray height) bi-weekly at each site to determine the variability in water quality parameters across sites.

Additionally, we collected water samples at each site during the growing season and analyzed the samples in the lab for total suspended solids (TSS) and chlorophyll-a

(Chl-a) using established protocols developed by the University of Rhode Island

Watershed Watch Program (www.uri.edu/ce/wq/ww). In addition, we used water quality data collected from a long-term monitoring buoy in Greenwich Bay

(http://www.narrbay.org), and from the URI Watershed Watch

(http://www.uri.edu/ce/wq/ww) water quality monitoring efforts in the coastal ponds to supplement our water quality data collections, and to provide a more comprehensive suite of water quality monitoring data.

Soil Characterization

We sampled the substrates of our oyster growth trial sites that represent both high energy [depositional shorelines (DS), washover fans (WFF, WFS), Platforms (P), and mainland beaches (MB)] and low energy [bay floor (BF), lagoon bottom (LB), and mainland cove (MC)] soil landscapes across two coastal ponds and two larger embayments (Figure 2.1) (Table 2.1). At the great group level, the soils classify as

Psammowassents (4 sites), Sulfiwassents (7 sites), and Haplowassents (1 site) (Table

2.1) (NRCS-SSURGO). In general, Psammowassents are associated with high-energy subaqueous soil landscapes characterized as having sandy textures throughout, and

low accumulations of organic matter when compared to low-energy soil landscapes.

Sulfiwassents are associated with low-energy depositional subaqueous soil landscapes and are dominated by fine textures, high organic matter accumulation, and higher total sulfides than Psammowassents. Haplowassents are typically associated with mainland beach subaqueous soil landscapes and can have a broad range of characteristics and morphologies.

We collected a soil core at each location using a vibracore (7.6 cm aluminum tube vibrated into the soil; Lanesky et al., 1979) or Macaulay peat sampler, depending on bottom type. Vibracores were sealed on both ends in the field, brought back to the lab, and kept at 4oC until being opened for description and characterization. Macaulay samples were described in the field, and samples were bagged and placed on ice to minimize sulfide oxidation, and transported to the lab and placed in the freezer at -

16oC. Soil characterization included particle size distribution (Gee and Bauder, 1986), bulk density (McVey et al., 2012), potential acidity (8-week incubation pH) (Soil

Survey Staff, 2010), and organic matter and CaCO3 contents (Rabenhorst, 1988; Heiri et al., 2001).

Statistical analysis

To characterize the variation of the oyster growth, percent survival, percent market size, and associated water quality data we conducted statistical analyses using

SigmaPlot 11.0 (Systat Software Inc., San Jose, CA). To explore the relationship between oyster growth, percent market size (% > 76 mm), and percent survival (% live oysters at end of 2nd year growth trials) across the different soil and environmental covariates we performed linear mixed effects modeling (LMM) and generalized linear

mixed effects modeling (GLMM) using R statistical environment 2.14.1 (R Core

Team, 2011), nlme (Pinheiro et al., 2014), and lme4 (Bates et al., 2015). For the linear mixed models, the mean oyster growth rate (mm/day) for each replicate was used as the response variable. For the generalized linear mixed models, the percent market size and percent survival for each replicate were used as the response variables. We included the site replicates as a random effect in all analyses to account for the nested study design and the correlation among replicates within each site. We investigated collinearity of the covariates using variance inflation factors, prior to inclusion for each candidate model.

We followed model selection and validation procedures outlined in Zuur et al.

(2007) and Zuur et al. (2009) to first identify the optimal random structure and subsequently identify the optimal fixed structure for each candidate model. We used

Akaike information criteria (AIC) to identify the optimal random structure and likelihood ratio testing, using maximum likelihood estimation, to determine the optimal fixed structure for the final candidate models. We obtained the final model parameter estimates using restricted maximum likelihood estimation for each of the candidate models. We validated the final models to check underlying assumptions by plotting residuals vs. fitted values to assess violations of homoscedasticity, and investigating histograms and QQ-plots to assess violations of normality (Zuur et al.,

2010).

RESULTS

Water Quality Characteristics

Several of the water quality parameters we monitored during the growth trials

(pH, salinity, dissolved oxygen (DO), temperature, total suspended solids (TSS), and

Chl-a concentration) were significantly different when comparing pooled data across the waterbodies using Kruskal-Wallis ANOVA on ranked data (Table 2.2). The only parameter that was not significantly different across at least one waterbody was pH (H

= 1.233, df = 3, P = 0.745) (Table 2.2). We observed the highest median pH in

Wickford Harbor (7.86) while the lowest median pH (7.64) was observed in Ninigret

Pond (Table 2.2). The median salinity (30.92 ppt) was significantly lower in Ninigret

Pond when compared to the other waterbodies we sampled (H = 12.036, df = 3, P =

0.007) (Table 2.2). We measured the lowest pooled median DO in Quonochontaug

Pond (7.24 mg/L), which was significantly lower than the other waterbodies we tested

(H = 10.58, df = 3, P = 0.014) (Table 2.2). Overall, the lowest DO we observed was in

Greenwich Bay (4.50 mg/L). Ninigret Pond had the highest median pooled surface water temperature overall compared to the other waterbodies tested (24.15 oC). The surface water temperature in Ninigret was significantly higher than Quonochontaug

Pond (H = 15.120, df = 3, P = 0,002), no other significant differences in surface water temperature were observed across the waterbodies sampled. Median chl-a was also variable across the different waterbodies. We observed significantly greater pooled median chl-a in the embayments when compared to the ponds using Kruskal Wallis

ANOVA on ranked data (Table 2.3). During the growth trials we measured the highest median chl-a concentration in Greenwich Bay (11.7 mg/L, σm = 0.81, n = 80), and the

lowest median concentration in Ninigret Pond (3.7 mg/L, σm = 0.21, n = 128), which was only slightly lower than the median concentration in Quonochontaug Pond (3.8 mg/L, σm =.0.19, n = 85) (Table 2.3). The median chl-a concentration in Wickford

Harbor was significantly greater than both the coastal ponds, however was also significantly lower than the median chl-a concentration we observed in Greenwich

Bay (Table 2.3).

Total suspended solids (TSS) was relatively low overall across all waterbodies we investigated. The median TSS in the ponds was significantly greater than the bays when compared using Kruskal Wallis ANOVA on ranked data (H = 9.855, df = 3, P =

0.020, α = 0.05). Both Ninigret Pond and Quonochontaug Pond had similar median pooled TSS values (21.2 mg/L, and 21.4 mg/L respectively). Both of the embayments had similar TSS values, with the median TSS in Wickford Harbor (17.0, mg/L) slightly higher than Greenwich Bay (16.5 mg/L), although both values were significantly lower than the coastal ponds (Table 2.4).

Oyster Growth

The mean oyster growth rate (mm/day) was highly variable across years and soil landscapes that we investigated ( = 0.19 mm/day, σ = 0.085). We observed the maximum growth rate (0.41 mm/day) during the 2011 season in Wickford Harbor at a

Psammowassent depositional shoreline site (WHDS). We observed the minimum growth rate (0.03 mm/day) at two different Sulfiwassent sites in Ninigret Pond

(NPLB, NPMC) in 2009 and 2013. Overall the pooled mean growth rate across years was significantly different ANOVA [F (4,42) = 8.384, P<0.001] (Fig. 2.2). The pooled mean growth rate in the two embayments (Greenwich Bay and Wickford Harbor) ( =

0.25 mm/day-1, σ = 0.075) was significantly greater when compared to the pooled mean growth rate in two coastal ponds ( = 0.16 mm/day-1, σ = 0.076 (Ninigret and

Quonochontaug) using one-way ANOVA [F (1,45) = 13.300, P <0.001] (Fig. 2.3).

One-way ANOVA comparing the pooled mean growth rate across the first and second growing seasons was also significant, with the first growing season mean growth rate

(0.241 mm/day-1) significantly greater than the second growing season mean growth rate (0.140 mm/day-1) [F (1,45) = 25.758, P = <0.001] (Fig. 2.4).

Results of the linear mixed model analysis indicate that different soil covariates including percent sand and percent organic matter within the A-horizon, soil classification at the Great Group level, Chl-a concentration, and oyster seed start size, were significant factors in explaining variability in oyster growth rate within the coastal ponds. The three candidate models, percent sand model (Table 2.6), percent organic matter model (Table 2.7), and the soil classification model (Table 2.8) indicate that each soil parameter investigated contributed significantly to explaining variation in mean oyster growth across the growth trials. AIC weights from the three candidate models suggest that the percent sand model had the greatest support (AICw = 0.81)

(Table 2.6) when compared to the percent organic matter model (AICw = 0.15) (Table

2.7) and soil the classification model (AICw =0.04) (Table 2.8). The effects plots, which represent graphical displays of the main effects, for each soil covariate demonstrate that percent sand within the A-horizon has a significant positive relationship with mean oyster growth, and percent organic matter within the A-horizon has a significant negative relationship with oyster growth (Figure 2.5). The effect plot for the soil classification candidate model indicates that Psammowassents and

Haplowassents were associated with above average oyster growth while Sulfiwassents were associated with below average growth (Figure 2.5).

Percent Market Size

The percent of oysters that reached market size (>76 mm) at the end of the 2nd year growth trials was highly variable among soil types (Table 2.9). The generalized linear mixed effects candidate models indicate that the soil covariates A-horizon soil organic matter, and soil classification at the great group level, were significant in explaining variation in percent market size across the growth trial sites within the coastal ponds (Tables 2.11 and 2.12). In addition to the soil covariates, the mean growth rate (mm/day) and year were also significant factors in explaining percent market size within each candidate model. AIC weights from the candidate models suggest that the soil classification model had the greatest support (AICw = 0.83)

(Table 2.10) when compared to the percent organic matter model (AICw = 0.18)

(Table 2.11). The effects plots for the two soil covariates demonstrate that percent organic matter within the A-horizon has a significant negative relationship with the percent market size (Figure 2.6). The effect plot for the soil classification candidate model indicates that Psammowassents were associated the greater mean predicted percentage of oysters reaching market size ( = 57%, 95% C.I. = 47%-67%) while the

Sulfiwassents were associated with the lowest mean predicted percentage of oysters reaching market size ( = 25%, 95% C.I.= 17%-35%) (Figure 2.6).

Oyster Survival

Oyster survival was highly variable across soils throughout the growth trials

(Table 2.12). The generalized linear mixed effects model indicates that the soil covariates, A-horizon soil organic matter, and percent sand were not significant in explaining variability in survival across the soil landscapes tested (P = 0.11) and

(P=0.17) respectively (Table 2.13). Soil classification at the great group level was also not significant at α = 0.05, however Psammowassents had greater survival when compared to Haplowassents and Sulfiwassents (P = 0.07) (Figure 2.7). The percent survival GLMM model also indicates that year, mean growth rate (mm/day) and oyster seed start size were significant factors in explaining percent survival (Table 2.13). The

GLMM percent survival effects plots demonstrate that the larger oyster seed (relative to the mean start size) within the growth trials was associated with greater percent survival. In addition, oysters that had above average growth rates were also associated with greater percent survival (Figure 2.7).

DISCUSSION

Although we observed significant differences in water quality parameters across some waterbodies including salinity, dissolved oxygen, temperature, and TSS, these differences are not likely to explain a majority of the variability in oyster growth and survival we observed during the growth trials. The water quality parameters we observed throughout the growth trials are within the published optimal ranges for juvenile and adult oysters, except for salinity, which is considered slightly higher than optimal (Table 5) (Galtsoff, 1964; Sellers & Stanley, 1984; Kennedy et al., 1996).

Median concentrations of total suspended solids (TSS) were significantly higher in the coastal ponds than the embayments (Table 2.4). A combination of wind driven circulation and relatively shallow depths in the coastal ponds may contribute to the higher TSS concentrations. Loosanoff and Tommers (1948), showed that pumping rates for eastern oysters are reduced between 50% - 87% with the addition of silt sized particles at water column concentration of 100 mg/L. In a more recent study, Suedel et al. (2014), indicated TSS concentrations as high as 500 mg/L had no impact on oyster survival, growth, and condition after seven days of exposure. It is unclear whether

TSS concentrations negatively influenced growth of oysters in the coastal ponds relative to the embayments; however, the low TSS concentrations we observed (< 22 mg/L) indicate that it is unlikely.

Chl-a concentration along with current velocity, have been identified as important parameters related to oyster growth (Grizzle and Lutz, 1989; Newell and Langdon,

1996; Rice, 1992; Rice and Pechenik, 1992). The median Chl-a concentrations we observed within the mid-bay region of Narragansett Bay were significantly greater,

and at least twice and nearly four times, in the case of Greenwich Bay, the median concentrations we observed within Ninigret and Quonochontaug ponds (Table 2.3).

The significant differences we observed in median chl-a concentration may explain some of the differences in growth between the embayment’s and the coastal ponds

(Table 2.3)(Figure 2.3). These results were anticipated, and are consistent with several research papers summarized by Oviatt, 2008 that document a gradient of nutrients and associated primary productivity extending from the Providence River Estuary in the upper bay to south to the opening of the West Passage at the entrance of Narragansett

Bay. However, the lack of hydrodynamic modeling data within the coastal ponds and embayments we investigated, except for recent modeling efforts in Greenwich Bay

(Balt, 2014; Kincaid et al. 2008) limits our understanding of the differences in seston flux within and across our study sites.

The relationships we observed between oyster growth and subaqueous soil characteristics within the coastal ponds is consistent with previous research that indicates coarse textured substrates support increased growth relative to fine textured substrates due to greater seston flux (Grizzle and Lutz, 1989; Rice, 1992; Rice and

Pechenik, 1992). The LMM and GLMM models indicate that oysters grown on coarse textured Psammowassents had greater growth rates and subsequently a greater proportion of oysters reached market size at the end of the 2nd year of the growth trials when compared to fine textured Sulfiwassents. The GLMM soil classification candidate model that compared the percentage of oysters reaching market size indicates on average 57% of oysters in the coastal ponds grown on course textured

Psammowassents reached market size at the end of the 2nd year growth trials compared

to 25% on average for oysters grown on fine textured Sulfiwassents. Therefore, being able to identify the presence and extent of these soil landscapes within the coastal ponds may be an important consideration for bivalve aquaculture and restoration site selection

Although limited to just one year of data from the growth trials, growth of second year seed oysters that were planted directly on the bottom showed no difference in growth when compared to oysters grown in elevated trays at each site. However, mortality of these oysters was nearly 100% when these direct on-bottom plots were established using first-year seed, which prevented comparisons. Additionally, we observed across all soil types that oysters grown directly on the bottom had less biofouling overall (it should be noted that this was not quantified).

Our data also indicate that oysters grown on Psammowassent soil landscapes had greater survival compared to oysters grown on Sulfiwassent soil landscapes within the coastal ponds, although not statistically significant in the model (P = 0.06) (Table

2.13). Additionally, larger, faster growing oyster seed had greater survival than smaller and slower growing oyster seed. Our data also show that oyster survival was highly variable across years, which may be associated with variable predation pressure, prevalence of disease, or other factors that we did not quantify. For example, we observed both Xanthid Mud crabs and oyster drills, which are identified as important predators of juvenile oysters, at variable densities across many sites.

The objectives of this research were to identify the best soils for oyster aquaculture and restoration activities within coastal ponds and embayments in southern New England. We conducted this research within dynamic estuarine

environments; therefore, we were unable to “control” and quantify all of the environmental conditions that may have influenced growth and survival, including prevalence of disease and predation pressure. During the course of the growth trials, our sites were exposed to two powerful hurricanes, Irene and Sandy, which impacted the southern New England coast resulting in lost replicates and some lost sites.

Additionally, due to regulatory mandates and availability, we were unable to use the same seed source across all of the growth trials, which introduced variability in growth and disease resistance depending on the seed source. Even with these difficulties, we were able to determine statistical differences in oyster growth and survival across different soils we investigated. Although the exact mechanisms for these differences are not evident, the observed patterns indicate subaqueous soil maps are a viable tool for coastal managers to identify priority sites for aquaculture development and restoration activities.

This study adds to the growing body of research that supports the use of a subaqueous soils mapping approach to characterize shallow subtidal substrates within coastal estuaries where detailed management interpretations are needed. Data based on the results our research were used to create a map of Quonochontaug Pond that outlines suitable locations for “on-the-bottom” aquaculture production methods which is the most prevalent production method currently used in the Rhode Island coastal ponds. Soil landscapes dominated by Psammowassents and Haplowassents that comprise 40 percent of the pond are identified as suitable areas for aquaculture development, whereas soil landscapes dominated by Sulfiwassents that comprise upwards of 60 percent of the pond are identified as not suitable for “on-the-bottom”

production methods because of slower growth rates, increased predation, and low bearing capacity (Fig. 2.8). In addition to soils data, planning maps that include areas of estuaries that have competing uses should be identified by the local community to aid in site selection process to reduce user conflict. These types of maps should be developed for other estuaries with heavy and variable use for managing shellfish aquaculture.

ACKNOWLEDGMENTS

This project was funded by The Nature Conservancy’s Global Marine Program and the Rhode Island Agricultural Experiment Station through funds projected for

Multistate Project NE-1038. We would like to thank Dr. Jose Amador, Dr. Adam

Smith, Dr. Gavino Puggioni for constructive suggestions and review of previous drafts. We would also like to thank Pamela Loring, Andrew Paolucci, Kristopher

Plante, Ashley Merlino, Ariana Pucci, and Mason Garfield from the URI Natural

Resources Science Department, Chris Littlefield, Boze Hancock, Steven Brown and

Dylan McNulty from The Nature Conservancy, well as Jim Turenne and Maggie

Payne from the USDA–NRCS for field and laboratory support during this project. We would also like to thank Dave Beutel from the Rhode Island Coastal Resources

Management Council for support throughout the duration of the project. This paper is a contribution of the Rhode Island Agricultural Experiment Station (no. 5416).

CITATIONS

Abbe, G.R., and B.W. Albright. 2003. An improvement to the determination of meat condition index for the eastern oyster, Crassostrea virginica (Gmelin 1791). J. Shellfish Res. 22:747-752.

Alleway, H.K., and S.D. Connell. 2015. Loss of an ecological baseline through the eradication of oyster reefs from coastal ecosystems and human memory. Con. Biol. 29:795-804.

Balt, C. 2014. Subestuarine circulation and dispersion in Narragansett Bay. Ph.D. Dissertation. Graduate School of Oceanography University of Rhode Island, Narragansett, RI.

Bates, D., M. Maechler, B. Bolker, and S. Walker. 2015. Fitting Linear Mixed-Effects Models Using lme4. Journal of Statistical Software, 67: 1-48.

Beck, M.W., R.D. Brumbaugh, L. Airoldi, A. Carranza, L.D. Coen, C. Crawford, O. Defeo, G.J. Edgar, B. Hancock, M.C. Kay, H.S. Lenihan, M.W. Luckenbach, C.L. Toropova, G. Zhang, and X. Guo. 2011. Oyster Reefs at Risk and Recommendations for Conservation, Restoration, and Management. BioScience 61: 107-116.

Beutel, D. 2014. Aquaculture in Rhode Island 2014 Annual Status Report. Rhode Island Coastal Resources Management Council. 10p. Wakefield, RI.

Boothroyd, J.C., N.E. Friedrich, and S.R. McGinn. 1985. Geology of microtidal coastal lagoons: Rhode Island. Mar. Geol. 63:35–76.

Bradley, M.P. 2001. Subaqueous soils and subtidal wetlands in Rhode Island. M.S. thesis. Dep. of Natural Resources Science, Univ. of Rhode Island, Kingston, RI.

Bradley, M.P., and M.H. Stolt. 2002. Evaluating methods to create a base map for a subaqueous soil inventory. Soil Sci. 167:222-228.

Bradley, M.P., and M.H. Stolt. 2003. Subaqueous soil–landscape relationships in a Rhode Island estuary. Soil Sci. Soc. Am. J. 67:1487–1495.

Bradley, M.P., and M.H. Stolt. 2006. Landscape-level seagrass–sediment relations in a coastal lagoon. Aquat. Bot. 84:121–128.

Bricker, S. B., T.L. Getchis, C.B. Chadwick, C.M. Rose, & J.M. Rose. 2016. Integration of ecosystem-based models into an existing interactive web-based tool for improved aquaculture decision-making. Aquaculture, 453: 135-146.

Brooks, W.K. 1996. The oyster. Johns Hopkins University Press, MD.

Burrell, V.G. 1986. Species profiles: Life histories and environmental requirements of coastal fishes and invertebrates (South Atlantic): American oyster. U.S. Fish Wildl. Serv. Biol. Rep. 82(11.57). U.S. Army Corp of Engineers. TR EL-82-4. 17p. Washington, DC.

Burge, C.A., M.C. Eakin, C.S. Friedman, B. Froliech, P.K. Hershberger, E.E. Hofmann, L.E. Petes, K.C. Prager, E. Weil, B.L. Willis, S.E. Ford, and C.D. Harvell. 2014. Climate change influences on marine infectious diseases: implications for management and society. Annu. Rev. Mar. Sci. 6:249-277.

Coen, L.D., R.D. Brumbaugh, D. Bushek, R. Grizzle, M.W. Luckenbach, M.H. Posey, S.P. Powers, and S.G. Tolley. 2007. Ecosystem services related to oyster restoration. Mar. Ecol. Prog. Ser. 341:303-307.

Davies, J.L. 1964. A morphogenetic approach to world shorelines. Z. Geomorphol. 8:127–142.

Demas, G.P. 1993. Submerged soils: a new frontier in soil survey. Soil Survey Horizons 34:44–46.

Demas, G.P., M.C. Rabenhorst, and J.C. Stevenson. 1996. Subaqueous soils: A pedological approach to the study of shallow-water habitats. Estuaries 19:229– 237.

Demas, G.P. 1998. Subaqueous soil of Sinepuxent Bay, Maryland. Ph.D. Dissertation, University of Maryland, College Park, MD.

Demas, G.P., and M.C. Rabenhorst. 1999. Subaqueous soils: Pedogenesis in a submersed environment. Soil Sci. Soc. Am. J. 63, 1250–1257.

Demas, G. P., and M.C. Rabenhorst. 2001. Factors of subaqueous soil formation: a system of quantitative pedology for submersed environments. Geoderma 102:189-204.

Eastern Oyster Biological Review Team. 2007. Status review of the eastern oyster (Crassostrea virginica). Report to the National Marine Fisheries Service, Northeast Regional Office. NOAA Tech. Memo. NMFS F/SPO-88. 105 p.

Erikson, L. 1998. Flushing times of Greenwich Bay, RI: Estimates from freshwater inputs. M.S. thesis. Dep. of Ocean Engineering, Univ. of Rhode Island, Narragansett, RI.

Food and Agriculture Organization of the United Nations. 2012. The State of World Fisheries and Aquaculture – 2010 (SOFIA), Rome.

Federal Geographic Data Committee. 1997. FGDC-STD-006: Soil geographic data standard. FGDC, Reston, VA.

Folger, D.W. 1972a. Texture and organic carbon content of bottom sediments in some estuaries of the United States. Geol. Soc. Am. Mem. 133:391-408.

Folger, D.W. 1972b. Characteristics of estuarine sediments of the United States. US Gov. Print. Off. No. 742. 94p.

Galtsoff, P.S. 1964. The American oyster Crassostrea virginica Gmelin. Fishery Bulletin Fish & Wildl Ser. 64:1-456.

Gee, G.W., and J.W. Bauder. 1986. Particle-size analysis. p. 383–411. In: A. Klute, editor, Methods of soil analysis. Part 1. 2nd ed. SSSA Book Ser. 5. ASA and SSSA, Madison, WI.

Grizzle, R. E., and R.A. Lutz. 1989. A statistical model relating horizontal seston fluxes and bottom sediment characteristics to growth of Mercenaria mercenaria. Mar. Biol. 102:95-105.

Grizzle, R.E., and P.J. Morin. 1989. Effects of tidal currents, seston, and bottom sediments on growth of Mercenaria mercenaria: results of a field experiment. Mar. Biol. 102:85-93.

Hayes, M. O. 1979. Barrier island morphology as a function of tidal and wave regime. P. 1-27. In: S.P. Leatherman, editor, Barrier islands from the Gulf of St. Lawrence to the Gulf of Mexico. Academic Press, New York, NY

Jenkins, J.B., A. Morrison, and C.L. MacKenzie, Jr. 1997. The molluscan fisheries of the Canadian Maritimes. p 15-44. In The History, Present Condition, and Future of the Molluscan Fisheries of North and Central America and Europe, Vol. 1. Atlantic and Gulf Coasts. (ed.) MacKenzie et al. U.S. Department of Commerce, NOAA Technical Report NMFS.

Jenny, H. 1994. Factors of soil formation: a system of quantitative pedology. Dover Pub. Inc. New York, NY.

Jespersen, J.L., and L.J. Osher. 2007. Carbon storage in the soils of a mesotidal Gulf of Maine estuary. Soil Sci. Soc. Am. J. 71:372–379.

Kennedy, S., and V. Lee. 2003. Greenwich Bay: An ecological history. Rhode Island Sea Grant Commun. Office, Univ. of Rhode Island, Narragansett, RI.

Kennedy, V.S., R.I.E. Newell and A.F. Able. 1996. The Eastern Oyster. Univ. MD Sea Grant Publications, College Park, MD. 734 pgs.

Kincaid, C., D. Bergondo, and, K. Rosenberger. 2008. The dynamics of water exchange between Narragansett Bay and Rhode Island Sound. p 301-324. In Science for Ecosystem-based Management. Springer. New York, NY.

Lanesky, D.E., B.W. Logan, R.G. Brown, and A.C. Hine. 1979. A new approach to portable vibracoring underwater and on land. J. Sediment. Petrol. 49:654–657.

Lee, V., and S. Olsen. 1985. Eutrophication and management initiatives for the control of nutrient inputs to Rhode Island coastal lagoons. Estuaries 8:191–202.

Loosanoff, V.L., and F.D. Tommers. 1948. Effect of suspended silt and other substances on rate of feeding of oysters. Science 107:69-70.

Mackenzie, C.L. 2007. Causes underlying the historical decline in eastern oyster (Crassostrea virginica Gmelin, 1791) landings. J. Shellfish Res. 26:927-938.

Mann, R., and E.N. Powell. 2007. Why oyster restoration goals in the Chesapeake Bay are not and probably cannot be achieved. J Shellfish Res. 26:905-917.

McVey, S., P.J. Schoeneberger, J. Turenne, M. Payne, D.A. Wysocki, and M.H. Stolt. 2012. Subaqueous soils (SAS) description. P 2-97 – 2-119. In: P.J. Schoeneberger et al., editors, Field book for describing and sampling soils. Version 3.0. Natl. Soil Surv. Ctr., Lincoln, NE.

Mercer, J., D. Leavitt, A. Cygler, M. LaFrance, D. Beutel, & J. McCann. 2015. Development of the Rhode Island state shellfish management plan and steps toward implementing recommendations. J. Shellfish Res. 34:664-664.

Millar, C.M., A.A.O. Aduomih, B. Still, and M.H. Stolt. 2015. Estuarine subaqueous soil organic carbon accounting: Sequestration and storage. Soil Sci. Soc. Am. J. 79:389-397.

Mitchell, S.B., T.C. Jennerjahn, S. Vizzini, and W. Zhang. 2015. Changes to processes in estuaries and coastal waters due to intense multiple pressures – an introduction and synthesis. Estuar. Coast. Shelf S. 156:1-6.

Munroe, D. M., Klinck, J. M., Hofmann, E. E., and E.N. Powell. 2014. A modelling study of the role of marine protected areas in metapopulation genetic connectivity in Delaware Bay oysters. Aquat. Conserv. 24:645-666.

Newell, R.I.E., and C.J. Langdon. 1996. Mechanisms and physiology of larval and adult feeding. P 185-229. In: V.S. Kennedy, R.I. Newell, and A.F. Eble, editors, The Eastern Oyster Crassostrea virginica. Maryland Sea Grant, College Park, MD.

NRCS. 2014. Web soil survey. Natl. Soil Surv. Ctr., Lincoln, NE. http://websoilsurvey.nrcs.usda.gov/ (accessed April 2014).

Osher, L.J., and C.T. Flannagan. 2007. Soil/landscape relationships in a mesotidal Maine estuary. Soil Sci. Soc. Am. J. 71:1323–1334.

Oviatt, C.A. 2008. Impacts of nutrients on Narragansett Bay productivity: A gradient approach. p 523-543, In: Desbonnet, A., and B.A. Costa-Pierce, editors, Science for Ecosystem-based Management – Narragansett Bay in the 21st Century. Springer New York, NY

Payne, M.K. 2007. Landscape-level assessment of subaqueous soil and water quality in shallow embayments in southern New England. M.S. thesis. Univ. of Rhode Island, Kingston, RI.

Payne, M.K., and J. Turenne. 2009. Mapping the “new frontier” of soil survey: Rhode Island’s mapcoast partnership. Soil Survey Horizons 50:86–89.

Powers, S. P., Peterson, C. H., Grabowski, J. H., & Lenihan, H. S. (2009). Success of constructed oyster reefs in no-harvest sanctuaries: implications for restoration. Mar. Eco. Prog. Ser. 389:159-170.

Powell, E. N., J. M. Klinck, K. Ashton-Alcox, E. E. Hofmann & J. M. Morson. 2012. The rise and fall of Crassostrea virginica oyster reefs: the role of disease and fishing in their demise and a vignette on their management. J. Mar. Res. 70:505– 558.

Pruett, C.M. 2010. Interpretations of estuarine subaqueous soils: Eelgrass restoration, carbon accounting, and heavy metal accumulation. M.S. thesis. Univ. of Rhode Island, Kingston, RI

R Core Team. 2011. R: A language and environment for statistical computing. R Found. Stat. Comput., Vienna.

Rabenhorst, M.C. 1988. Determination of organic and carbonate carbon in calcareous soils using dry combustion. Soil Sci. Soc. Am. J. 52:965–969.

Rabenhorst, M.C., and M.H. Stolt. 2012. Subaqueous soils: Pedogenesis, mapping, and applications. p 173-204. In: H. Lin, editor, Hydropedology: Synergistic integration of soil science and hydrology. Academic Press, San Diego, CA.

Rice, M.A. 1992. The northern quahog: the biology of Mercenaria mercenaria. Rhode Island Sea Grant, RIU-B-92-001. ISBN 0-938412-33-7. 69 p.

Rice, M. A., and J.A. Pechenik. 1992. A review of the factors influencing the growth of the northern quahog, Mercenaria mercenaria Linnaeus, 1758. J. Shellfish Res. 11:279-287.

Rice, M. A., Valliere, A., and A. Caporelli. 2000. A review of shellfish restoration and management projects in Rhode Island. J. Shellfish Res. 19:401-408.

Rhode Island Department of Environmental Management. 2015. Notice of polluted shellfish grounds. RIDEM Office of Water Resources. Providence, RI. 45p.

Rothschild, B. J., Ault, J. S., Goulletquer, P., and M. Heral. 1994. Decline of the Chesapeake Bay oyster population: a century of habitat destruction and overfishing. Mar. Eco. Prog. Ser. 111:29-39.

Salisbury, A.R. 2010. Developing subaqueous soil interpretations for Rhode Island estuaries. M.S. thesis. Univ. of Rhode Island, Kingston, RI.

Salisbury, A., and M.H. Stolt. 2011. Estuarine subaqueous soil temperature. Soil Sci. Soc. Am. J. 75:1584–1587.

Schulte, D. M., and R.P. Burke. 2014. Recruitment Enhancement as an Indicator of Oyster Restoration Success in Chesapeake Bay. Ecol. Rest. 32:434-440.

Seitz, R. D., H. Wennhage, U. Bergström, R.N. Lipcius, and T. Ysebaert. 2014. Ecological value of coastal habitats for commercially and ecologically important species. ICES J. Mar. Sci. 71:648-665.

Sellers, M.A. and J.G. Stanley. 1984. Species profiles: life histories and environmental requirements of coastal fishes and invertebrates (North Atlantic): American oyster. U.S. Fish. Wildl. Serv. FWS/OBS-82/11.23. U.S. Army Corps of Engineers, TR EL-82-4. 15 pgs.

Shumway, S. E., C. Davis, R. Downey, R. Karney, J. Kraeuter, J. Parsons, & G. Wikfors. 2003. Shellfish aquaculture–in praise of sustainable economies and environments. World Aqua. 34:8-10.

Shumway, S.E. 1996. Natural environmental factors. p 467-513. In: V.S. Kennedy, R.I.E. Newell and A.F. Eble, editors. The Eastern Oyster Crassostrea virginica. Maryland Sea Grant College, University of Maryland, College Park, MD.

Simonson, R.W. 1959. Outline of a generalized theory of soil genesis. Soil Sci. Soc. Am. J. 23:152-156.

Soil Survey Staff. 1999. Soil Taxonomy. U.S. Dept. Agric. Handbook No. 436, second ed. US Govt. Printing Office, Washington, DC.

Soil Survey Staff. 2010. Keys to Soil Taxonomy. 11th ed. US Gov. Print. Office, Washington, DC.

Soniat, T. M., Cooper, N., Powell, E. N., Klinck, J. M., Abdelguerfi, M., Tu, S. Tu, R. Mann, and P.D. Banks. 2014. Estimating Sustainable Harvests of Eastern Oysters, Crassostrea virginica. J. Shellfish Res. 33:381-394.

Still B.M., and M.H. Stolt. 2015. Subaqueous soils and coastal acidification: A hydropedology perspective with implications for calcifying organisms. Soil Sci. Soc. Am. J. 79:407–416.

Stolt, M., M. Bradley, J. Turenne, M. Payne, and E. Scherer. 2007. From land to shallow water: The evolution of soil mapping in the MapCoast Partnership. 41N 4:4.

Stolt, M., M. Bradley, J. Turenne, M. Payne, E. Scherer, G. Cicchetti, et al. 2011. Mapping shallow coastal ecosystems: A case study of a Rhode Island lagoon. J. Coast. Res. 27:1–15.

Stolt, M.H., and M.C. Rabenhorst. 2011. Subaqueous soils. p 36-1–36-14 In: P.M. Huang et al., editors, Handbook of Soil Science. 2nd ed. CRC Press, Boca Raton, FL.

Suedel, B.C., J.U. Clarke, J. Wilkens, C.H. Lutz, and D.G. Clarke. 2014. The effects of a simulated suspended sediment plume on eastern oyster (Crassostrea virginica) survival, growth, and condition. Estuaries and Coasts 38:578-589.

USDA Census of Aquaculture. Vol 3. Special Studies Part 2. 2014. AC-12-SS-2. United States Department of Agriculture, National Agricultural Statistics Service. 98.p

Zu Ermgassen, P. S., Spalding, M. D., Grizzle, R. E., and, R.D. Brumbaugh. 2013. Quantifying the loss of a marine ecosystem service: filtration by the eastern oyster in US estuaries. Estuar. Coast. 36:36-43.

Zuur, A.F., E.N. Ieno, and C.S. Elphick. 2010. A protocol for data exploration to avoid common statistical problems. Methods Ecol. Evol. 1:3–14.

Zuur, A.F., E.N. Ieno, and G.M. Smith. 2007. Analysing ecological data. Springer, New York, NY

Zuur, A.F., E.N. Ieno, N. Walker, A.A. Saveliev, and G.M. Smith. 2009. Mixed effects models and extensions in ecology with R. Springer, New York, NY

TABLES

Table 2.1. Taxonomic classification and general characteristics for the surface horizon of the subaqueous soils we used for the oyster growth trials. Data obtained from Natural Resources Conservation Service (NRCS) Soil Survey Geographic Database SSURGO. Subgroup Landscape Clay Sand OM Series Site Classification Position (%) (%) (%) WH-BF Fluventic NP-LB Bay Floor Pishagqua 17 18 14 Sulfiwassents QP-LB Lagoon Bottom NP-MB Fluventic GB-DS Depositional Rhodesfolly 1 98 1.5 Psammowassents GB-P Shoreline

Thapto-histic Billington NP-MC Mainland Cove 17 18 14 Sulfiwassents

Fluventic Depositional Massapog WH-DS 0 97 1 Psammowassents Shoreline NP-WFF Sulfic NP-WFS Nagunt Washover Fan 1 98 1 Psammowassents QP-WFF QP-WFS Aeric Napatree QP-MB† Mainland Beach 0 99 1 Haplowassents † Site not included in mixed model analysis.

Ninigret Pond Median 7.7 30.9 a 7.6 ab 24.2 a St Dev 0.3 1.9 1.5 3.8 Min 7.1 26.8 4.7 9.7 Max 8.5 34.9 12.1 27.5 N 106 82 90 101

Quonochontaug Pond Median 7.7 31.8 b 7.2 a 22.0 b St Dev 0.3 2.9 1.0 2.8 Min 7.0 21.5 4.7 14.6 Max 8.16 34.64 8.92 25.40 N 65 45 53 61

Greenwich Bay Median 7.8 31.8 ab 7.6 ab 22.8 ab St Dev 0.4 1.6 1.5 3.5 Min 6.8 28.1 4.5 16.1 Max 8.1 33.5 10.3 25.4 N 20 20 20 20

Wickford Harbor Median 7.9 31.9 ab 8.1 b 23.3 ab St Dev 0.4 1.8 1.3 5.3 Min 7.1 27.7 5.5 9.4 Max 8.5 34.1 9.6 28.2 N 33 30 30 33

H 1.233 12.036 10.582 15.120 df 3 3 3 3 P 0.745 0.007 0.014 0.002

Median Waterbody n s Chl-a (mg/L) m Ninigret Pond 128 3.7 c 0.21 Quonochontaug Pond 85 3.8 c 0.19 Greenwich Bay 80 11.7 a 0.81 Wickford Harbor 27 6.1 b 0.85

Table 2.4. TSS data collected during the oyster growth trials (2008 – 2009, 2011 – 2013). Significant differences among locations were determined using Kruskal-Wallis ANOVA on Ranks (H). Letters represent significant differences when comparing median TSS across water bodies using Dunn’s multiple comparisons (H = 9.855, df = 3, P = 0.020, α = 0.05). Dashed line separates coastal ponds from embayments.

Median TSS Waterbody n s (mg/L) m Ninigret Pond 89 21.2 a 16.9

Quonochontaug Pond 60 21.4 a 11.1

Greenwich Bay 20 16.5 b 8.6

Wickford Harbor 29 17.0 b 6.7

Table 2.5. Environmental parameters considered ideal for juvenile – adult stage eastern oysters from the literature.

Environmental Parameters Juvenile - Adult

Temperature (oc) 20.0 - 30.0

Salinity (ppt) 14.0 - 28.0

Dissolved Oxygen (mg/L) > 4.0 pH 6.75 - 8.75

Turbidity (mg/L) < 750

Water flow (cfs) > 10 Adapted from RI Shellfish Management Plan (2014) (Galtsoff, 1964; Sellers & Stanley, 1984; Kennedy et al., 1996)

Table 2.6. Linear mixed effects model results for the oyster growth vs. percent sand candidate model using restricted maximum likelihood estimation (REML). Data for this model are from the oyster growth trials within the two coastal ponds we investigated (Ninigret Pond and Quonochontaug Pond). The continuous variables in the model were standardized to improve model convergence. Variables Value SE DF T-value P-value Percent Sand Candidate Model Fixed Effects Intercept 0.00 0.074 74 0.00 1.000 Percent Sand (A horizon) 0.279 0.076 22 3.68 0.0013 Chl-a -0.234 0.078 74 -3.01 0.0035 Avg. start size -0.525 0.077 74 -6.82 0.0000 Random Effect (intercept) Std. Dev. Residual Error Rep 0.000026 0.729 Model AIC: 232.7 Model ∆AIC: 0 Model AICw: 0.81

Table 2.7. Linear mixed effects model results for the oyster growth vs. percent organic matter candidate model using restricted maximum likelihood estimation (REML). Data for this model are from the oyster growth trials within the two coastal ponds we investigated (Ninigret Pond and Quonochontaug Pond). The continuous variables in the model were standardized to improve model convergence. Variables Value SE DF T-value P-value Percent Organic Matter Candidate Model Fixed Effects Intercept 0.00 0.076 74 0.00 1.000 Percent Organic Matter -0.239 0.077 22 -3.13 0.0048 Chl-a -0.293 0.078 74 -3.74 0.0004 Avg. start size -0.515 0.078 74 -6.58 0.0000 Random Effect (intercept) Std. Dev. Residual Error Rep 0.000051 0.758 Model AIC: 249.47 Model ∆AIC: 3.33 Model AICw: 0.15

Table 2.8. Linear mixed effects model results for the oyster growth vs. Soil Classification candidate model using restricted maximum likelihood estimation (REML). Data for this model are from the oyster growth trials within the two coastal ponds we investigated (Ninigret Pond and Quonochontaug Pond). The continuous variables in the model were standardized to improve model convergence. Variables Value SE DF T-value P-value Soil Classification: Great Group Model Fixed Effects Intercept 0.291 0.234 74 1.23 0.218 Psammowassents -0.129 0.265 21 -0.49 0.63 Sulfiwassents -0.546 0.265 21 -2.06 0.052 Chl-a -0.236 0.082 74 -2.89 0.0051 Avg. start size -0.525 0.080 74 -6.55 0.0000 Random Effect (intercept) Std. Dev. Residual Error Rep 0.131 0.758 Model AIC: 252.4 Model ∆AIC: 6.26 Model AICw: 0.04

Table 2.9. Percent market size (% > 76 mm) for oysters grown in standard aquatrays after two growing seasons. The numbers in parentheses for seed 3 represent oysters that were grown directly on the bottom during the 2nd growth year in 2013.

Seed 1 Seed 2 Seed 3 Site Yr-2 (2009) Yr-2 (2012) Yr-2 (2013)

NP-LB 1 ‡ 19 (17)

NP-MC 47 ‡ 0 (4)

NP-WF 44 † †

NP-WFS 83 97 60 (68)

QP-LB 24 56 10 (30)

QP-MB 62 ‡ 24 (12)

QP-WF 62 † †

QP-WFS 62 73 26 (29)

WH-BF † 92 88 (90)

WH-DRC † 94 67 (67)

WH-DS † 97 70 (61)

GB-DS † ‡ ‡

GB-P † ‡ 17 (17) † Sites were not established during this year of the growth trials ‡ Sites were lost due to storm events or vandalism.

Table 2.10. Generalized linear mixed model results for fixed effects and random intercept terms included in the oyster percent market size vs. Soil Classification candidate model. Data for this model are from the second year oyster growth trials within the two coastal ponds we investigated (Ninigret Pond and Quonochontaug Pond). Variables Value SE Z-value P-value Soil Classification: Great Group Model Fixed Effects Intercept -1.802 0.568 -3.174 0.0015 Psammowassents 0.915 0.480 1.905 0.057 Sulfiwassents -0.461 0.507 -0.911 0.362 Growth rate 10.971 2.040 5.378 0.000 Year 2012 1.248 0.223 5.591 0.000 Year 2013 -0.998 0.175 -5.702 0.000 Random Effect (intercept) Std. Dev. Rep 0.6369 Model AIC: 273 Model ∆AIC: 0.0 Model AICw: 0.83

Table 2.11. Generalized linear mixed model results for fixed effects and random intercept terms included in the oyster percent market size vs. Percent Organic Matter candidate model. Data for this model are from the second year oyster growth trials within the two coastal ponds we investigated (Ninigret Pond and Quonochontaug Pond). Variables Value SE Z-value P-value Percent Organic Matter Model Fixed Effects Intercept -0.921 0.376 -2.449 0.0143 Percent organic matter -0.128 0.038 -3.333 0.0009 Growth rate 10.438 2.090 4.994 0.0000 Year 2012 1.271 0.224 5.665 0.0000 Year 2013 -1.053 0.175 -6.035 0.0000 Random Effect (intercept) Std. Dev. Rep 0.752 Model AIC: 276.1 Model ∆AIC: 3.1 Model AICw: 0.18

Table 2.12. Percent survival at the end of the growing season for each site and seed source. Percent survival was not determined at the end of the 2008 growing season.

Seed 1 Seed 2 Seed 3 Site Yr-2 (2009) Yr-1 (2011) Yr-2 (2012) Yr-1 (2012) Yr-2 (2013)

NP-LB 15 ‡ ‡ 20 87

NP-MC 25 51 ‡ 56 52

NP-WF 23 † † † †

NP-WFS 21 43 55 80 79

QP-LB 29 50 44 57 53

QP-MB 37 ‡ ‡ 42 75

QP-WF 50 † † † †

QP-WFS 37 18 83 57 91

WH-BF † 38 68 55 91

WH-DRC † 27 57 56 78

WH-DS † 40 12 38 65

GB-DS † ‡ ‡ 13 ‡

GB-P † 37 ‡ 20 18 † Sites were not established during this year of the growth trials ‡ Sites were lost due to storm events or vandalism.

Table 2.13. Generalized linear mixed model (GLMM) results for fixed effects and random intercept terms included in the oyster percent survival model. Data for this model are from the second year oyster growth trials within the two coastal ponds we investigated (Ninigret Pond and Quonochontaug Pond). The continuous variables in the model were standardized to improve model convergence. Variables Value SE Z-value P-value Oyster Percent Survival Model Fixed Effects Intercept -1.345 0.319 -4.218 0.0000 Year 2011 0.348 0.429 0.810 0.418 Year 2012 1.314 0.346 3.800 0.0000 Year 2013 2.589 0.267 9.686 0.0000 Growth rate 0.375 0.137 2.742 0.006 Psammowassents 0.577 0.316 1.828 0.067 Sulfiwassents 0.078 0.310 0.252 0.801 Avg. start size 0.286 0.150 1.905 0.057 Random Effect (intercept) Std. Dev. Rep 0.000 Eps (observation level error term) 0.818 Model AIC: 752.8

FIGURES

Figure 2.1. Map of study sites in Rhode Island. Inset maps represent locations of oyster growth trials in Greenwich Bay, Wickford Harbor, Quonochontaug Pond and Ninigret Pond.

Figure 2.7. Covariates that we investigated within the percent survival generalized linear mixed model. Gray bands and error bars represent 95% confidence intervals. A) Soil Taxonomy Great Group covariate showing the predicted percent survival across soil taxonomy at the great group level. B) Year covariate showing the predicted percent survival across oyster growth trial years. C) Oyster start size covariate showing the predicted percent survival across avg. oyster start size (mm) at the beginning of the growth trials. D) Oyster growth rate (mm/day) covariate showing the predicted percent survival across observed oyster growth rates during the growth trials.

Sulfiwassent Soils

Psammowassent Soils

Haplowassent Soils

0 0.25 0.5 1 Kilometers ¹

Figure 2.8. Subaqueous soil interpretation for oyster aquaculture development using on-the-bottom methods based on results of growth trials within Quonochontaug Pond. Psammowassent and Haplowassent soils depicted in yellow and pink represent areas that may provide increased growth rates relative to the Sulfiwassent soils depicted in red.

MANUSCRIPT 3 - SPATIAL PLANNING FOR OYSTER AQUACULTURE:

APPLICATION OF SUBAQUEOUS SOIL MAPS

Manuscript will be sent to Soil Science Society of America Journal for publication

Brett M. Still1, Mark H. Stolt1

1Department of Natural Resources Science, University of Rhode Island

1 Greenhouse Road, Kingston, RI 02881

ABSTRACT

Expansion of oyster aquaculture has reversed this trend, creating a growing market for oysters as a food resource. This growth however, is being constrained by a number of issues, particularly public use and access to the coastal zone (user conflicts) and identification of productive locations for aquaculture lease development. In an effort to reconcile these issues, we developed a GIS-based support tool to couple subaqueous soils data that identifies submerged soil landscapes that support productive oyster aquaculture with spatial data of non-compatible uses that identifies aquaculture restriction zones. Our studies focused on the Rhode Island coastal salt ponds region, which has seen significant increases in aquaculture development over the last decade.

The goal was to develop a support tool that coastal managers and regulators could use as a component of comprehensive shellfish aquaculture management planning. We found that between 43 to 70% of the coastal salt ponds represent non-compatible uses including boating and navigation, submerged aquatic vegetation, and recreational shellfishing. Of the remaining available area, soil landscapes that can support productive on-bottom culture ranges from 2% - 34%, depending on the salt pond.

Currently, 2% (95.3 ac) of the salt ponds are used for aquaculture, leaving 3% (143 acres) available for lease development, given current regulations. The subaqueous soils approach we tested provides the needed resolution for aquaculture lease

development, as well as useful information for restoration site selection. As the extent of subaqueous soil survey continues to expand, subaqueous maps will increasingly be available as a planning tool for coastal managers supporting the growth of the industry, while managing user conflict.

INTRODUCTION

The coastal zone in the U.S. accommodates 53% of the US population, yet only

17% of the total land area (Crossett et al., 2004). Numerous studies have documented ecosystem responses related to this increase in development including nutrient enrichment, loss of eelgrass, increased macro-algal cover, decline in fisheries resources, reduced water quality, and increased sedimentation (Bowen and Valiela,

2001; Deegan et al., 2002; Nixon, 1993; Nixon, 1995; Nixon et al., 2001; Short et al.,

1996). These stressors including a long history of exploitative harvest have led to significant declines of wild oyster populations in many estuaries in the U.S. and around the world. Many of these populations have become functionally extinct, no longer providing ecosystem services within these systems (Alleway & Connell, 2015;

Beck et al., 2011; Halpern et al., 2008; Jackson et al., 2001; Mackenzie, 2007;

Mitchell et al., 2015; Rothschild et al., 1994; Seitz et al., 2014; Zu-Ermgassen et al.,

2013).

The development and expansion of oyster aquaculture has reversed this trend, creating a growing market for oysters throughout the U.S. and abroad. Data from the

FAO indicates that aquaculture of shellfish accounts for greater than 80% of total global shellfish production, with oyster aquaculture accounting for nearly 95% of global oyster production (Cochrane et al., 2009; Clements and Chopin., 2016). In the

United States, the total value of aquaculture products (fish and shellfish) has increased

26% since 2005 with a current production value of $1.37 billion (USDA, 2013). Over that same time, total clam production has increased 45%, and total oyster production has increased 75%, with a combined bivalve aquaculture value of $303 million

(USDA, 2013). Bivalve aquaculture in the New England states (ME, NH, MA, RI, and

CT) account for approximately 15% of the total U.S. shellfish production with the latest estimated harvest value of $47 million (Table 1). On a dollar per acre basis, shellfish aquaculture dwarfs returns from traditional agriculture in Rhode Island

(USDA, 2015). For example, average production yields of $25,242 dollars per leased acre were achieved in RI in 2014 (Beutel, 2014). Continued expansion of the shellfish aquaculture industry is needed to meet growing demand; however, growth will be constrained by access to suitable locations, poor water quality, and user conflict issues.

At the Federal level, the United States is taking action with the National Ocean

Policy Implementation Plan (National Ocean Council, Washington, DC; 2013), and the NOAA National Shellfish Initiative (NOAA 2011) to encourage federal, state and local management agencies, the fisheries community, and NGOs to develop comprehensive ecosystem based management plans in an effort to increase aquaculture production, create jobs, and recover ecosystem services that have been lost (Knapp and Rubino, 2016). Such plans have recently been developed along the

Pacific coast in Washington and California, as well as Maine and Maryland along the

Atlantic coast (Pacific Coast Shellfish Growers Association, 2011; Knapp and Rubino

2016). In 2014, the state of RI completed a comprehensive statewide Shellfish

Management Plan (SMP) focused on commercial harvest, aquaculture, restoration and recreational fisheries (SMP, 2014). The SMP process involved a diverse group of stakeholders including; state regulatory agencies RIDEM and CRMC, non-profit conservation organizations, the wild harvest and aquaculture industries, the scientific

community, and recreational users, with the goal to find consensus across a broad range of shellfish management issues.

Acquiring an aquaculture lease within many states, including Rhode Island, is a multi-step process. State and local regulators consult with the applicant as well commercial fisheries representatives, and other stakeholders to arrive at a consensus decision to support or deny the proposed lease site (East Coast Shellfish Growers

Association, 2014; CRMC, 2008). The goal of this process is to identify possible user conflicts and work with the applicant to site the proposed aquaculture lease within an area that is not heavily used to reduce incompatible use conflicts (CRMC, 2008).

Through the SMP ecosystem management plan process, coastal ecosystem managers have identified a critical need for a support tool for making use and management decisions for shallow coastal systems at an ecosystem scale for managing aquaculture development and potential user conflict. One of these recommendations suggests the use of coastal resource area maps that identify “best sites for aquaculture”

(SMP 2014). The SMP recommendation also states that these maps should include ecological characteristics that support good shellfish growth as well as identify areas that support conflicting use activities such as boating, fishing, etc., as a way to limit conflicting uses with continued aquaculture development (SMP 2014).

Recent advances in subaqueous soils mapping standards, and the corresponding ecological relationships, demonstrate the value of subaqueous soil maps as a support tool for coastal use management (Stolt and Rabenhorst, 2011; Stolt et al., 2011;

Rabenhorst and Stolt, 2012). Several recent studies have shown subaqueous soil survey information can be applied across a range of coastal management applications

including submerged aquatic mapping and restoration, soil carbon accounting, coastal acidification planning, and shellfish management (Bradley and Stolt, 2006; Jespersen and Osher, 2007; Payne, 2007; Millar et al., 2015; Still & Stolt 2015; Still et al., in preparation; Salisbury, 2010). Further, the NOAA Coastal and Marine Ecological

Classification Standard (CMECS) recommends that a subaqueous soils survey and classification approach be used in shallow estuarine waters where detailed management and interpretations are required (FDGC, 2012).

Soil survey and interpretations have been developed and used for determining best use and management of terrestrial landscapes for over 100 years in the U.S. (Soil

Survey Division Staff, 1993). Early on, these interpretations were developed for agricultural and forest management activities. By the 1950’s, soil interpretations were also being developed for engineering applications including road building and foundations. More recently, interpretations have been developed for, citing on-site septic systems, crop yields, and recreational activities among others. (Soil Survey

Division Staff, 1993). Over the last decade estuarine subaqueous soil interpretations have been established (Stolt and Rabenhorst, 2011; Rabenhorst and Stolt, 2012). This research is aimed at illustrating the use of subaqueous soil surveys to develop use and interpretations for aquaculture development and planning.

In this study, we developed a GIS-based support tool to couple subaqueous soils data that identifies soil landscapes that support productive oyster aquaculture with spatial data of non-compatible uses that identifies aquaculture restriction zones. Our studies focused on the coastal ponds region of southern Rhode Island because although the coastal salt ponds comprise roughly 5% of the total Rhode Island waters,

100

42% of the existing public trust lands leased for aquaculture production statewide are found here (Beutel, 2014). In addition, the coastal salt ponds in Rhode Island are part of expanding subaqueous soil mapping efforts along the Atlantic coast of the U.S. This mapping will provide needed spatial resource data for management of shallow coastal estuaries throughout the region. Our goal is to develop a GIS-based decision support tool that incorporates: (1) detailed habitat scale subaqueous soil maps with (2) non- compatible use spatial data to identify areas within the Rhode Island coastal salt ponds that could support aquaculture development and shellfish restoration while reducing potential user conflict. These data are supported by soil-based studies of oyster shell dissolution (Still and Stolt, 2015) and long-term oyster growth (Still et al. in preparation) that form the basis for the oyster aquaculture production and restoration interpretations.

101

METHODOLOGY

Study Area

We developed this aquaculture decision support tool for the coastal salt pond region along the south coast of Rhode Island that includes coastal salt ponds within the towns of Narragansett, South Kingstown, Charlestown, and Westerly (Figure 3.1). We selected this region because the Mapcoast Partnership and Laboratory of Pedology and

Soil Environmental Science (LPSES) have conducted extensive subaqueous soils mapping throughout the region and has complete subaqueous soils data for the coastal salt ponds. These subaqueous soils data are available through the State of Rhode

Island Geographic Information System (RIGIS) (http://www.edc.uri.edu/rigis/) and

NRCS Web Soil Survey (http://websoilsurvey.sc.egov.usda.gov/App/HomePage.htm).

The LPSES and other researchers have conducted studies investigating oyster and quahog growth across different soil landscapes within the ponds as well as habitat suitability modeling for oyster restoration activities (Brown et al., 2013; Salisbury,

2013; Still et al., in preparation). In addition, the coastal pond region has been the focus of ecological carrying capacity modeling of bivalve aquaculture (Byron et al.,

2011), and part of a state level Coastal Pond Region Special Area Management Plan

(Ernst et al., 1999).

Five coastal salt ponds within the region currently support bivalve aquaculture including Point Judith, Potter, Ninigret, Quonochontaug, and Winnapaug (Figure 3.1).

Ninigret and Point Judith Ponds are the largest of the salt ponds, having surface areas of 1,601 ac and 1,580 ac respectively. The surface area of Quonochontaug Pond is approximately 760 ac. Winnapaug and Potter Ponds are the smallest of the coastal salt

102

ponds, occupying 472 ac and 360 ac respectively (Table 3.1). We derived the surface area for each pond using available RIDEM Office of Water Resources GIS data that identifies portions of each pond that are approved, conditionally approved, or prohibited for the taking of shellfish (RIDEM, 2015). These designations are based on

Interstate Shellfish Sanitation Commission water quality criteria, administered by

RIDEM Office of Water Resources.

In general, the coastal ponds are described as micro-tidal, mixed energy, wave dominated coastal lagoon estuarine systems separated from Block Island Sound by barrier beaches with maintained permanent breachways for navigational access

(Davies, 1964; Hayes, 1979, Boothroyd, 1985; Ernst, 1999). The submerged soil landscapes within these systems include lagoon bottom (LB), Flood Tide Deltas

(FTD), Mainland Cove (MC), submerged mainland beach (SMB), Washover Fan Flat

(WFF), and Washover Fan Slope (WFS), which represent a range of depositional and high energy environments (Bradley and Stolt, 2003).

Subaqueous Soils

We aggregated the subaqueous soils data at the Great Group soil classification level using Soil Taxonomy to group soils with similar overall characteristics within these systems (Soil Survey Staff, 1999). At this taxonomic level, the soils within the salt ponds are mapped as Sulfiwassents, Psammowassents, or Haplowassents. The relative proportion of these soils is variable across each individual pond; however, the overall pattern is similar.

Sulfiwassents form in fine-silty marine and estuarine deposits with high organic matter content, high sulfide accumulation, low bulk density, and low bearing capacity

103

(Soil Survey Staff, 2016). Sulfiwassents are associated with depositional basins including LB soil landscapes. Of the total acreage across the salt ponds, 61% (2,932 ac) of the soils are mapped as Sulfiwassents. These soils comprise the majority of the area within four of the five coastal salt ponds, including 71% of Point Judith Pond

(Table 3.3).

Psammowassents formed from sandy marine deposits. These soils have higher bulk density, lower soil organic matter, and lower sulfide accumulation when compared to Sulfiwassents. Psammowassents are associated with WFF and WFS soil landscapes, as well as FTD soil landscapes within coastal lagoons (NRCS, 2016).

These soil landscapes are associated with higher energy environments than depositional basins, and are maintained via barrier beach wash-over storm events, eolian deposition, and tidal flooding. Thirty-three percent (1,595 ac) of the subaqueous soils across the salt ponds are mapped as Psammowassents (Table 3.3).

Psammowassents occupy 50% of Winnapaug, 37% of Ninigret, 34% of Potter, and

28% of both Point Judith and Quonochontaug Ponds (Table 3.3).

Haplowassent soils are associated with MB and MC soil landscapes adjacent to glaciated upland shorelines and have a range of characteristics and soil morphologies.

Only five percent (245 ac) of the subaqueous soils across the salt ponds are mapped as

Haplowassents (Table 3.3). Fourteen percent of the soils in Quonochontaug are mapped as Haplowassents. While only 11% of Winnapaug, 4% in Ninigret, and 1% of

Point Judith are mapped as Haplowassents (Table 3.3). Haplowassents were not mapped in Potter Pond, meaning areas that did not meet the minimum mapping unit of

0.5 acres.

104

Geographic Information Systems

We generated a suite of spatial data for the coastal ponds to conduct an analysis of existing use activities that may conflict with aquaculture lease development using

ESRI ArcGIS version 10.2.2 (ESRI, 2014). We compiled existing spatial data from the

State of Rhode Island Geographic Information System (RIGIS, 2016)

(http://www.edc.uri.edu/rigis/) including, RIDEM Office of Water Resources shellfish classification areas, submerged aquatic vegetation (2009, 2012), subaqueous soils data, and basemap orthophoto imagery from the RI Environmental Data Center

ArcGIS map server. We also compiled the most up-to-date existing bivalve aquaculture lease areas for Rhode Island from the Northeast Ocean Data Portal

(http://www.northeastoceandata.org/) (current as of May 2014). In addition to the

RIGIS data, we incorporated recreational shellfish zones within the coastal pond region that were identified through a series of stakeholder meetings during the SMP process. We created a suite of spatial data for the coastal ponds that identified navigation centerlines, mooring/anchorage areas, docks/piers that support boating, and other water dependent uses (Table 3.2). We aggregated these user group data into a single spatial data layer within ESRI ArcMap 10.2.2 (ESRI, 2014) to identify areas of the pond that may preclude establishment of leases for bivalve aquaculture due to potentially conflicting uses (Table 3.2). Once established, we calculated the area of each pond occupied by this “aquaculture restriction zone” (ARZ) to estimate the proportion of the total pond area that might be excluded from aquaculture development due to conflicting uses.

105

In addition to these spatial data, we integrated the results of two studies conducted within the coastal pond region that used subaqueous soils data to characterize pH variability and oyster shell dissolution (Still and Stolt., 2015) and oyster growth experiments to identify soil landscapes that exhibit increased oyster growth rates within these systems (Still et al., in preparation). We linked these data with the mapped soils to support interpretations of the best soils for expansion of on- the-bottom grow out, floating aquaculture production methods, and oyster restoration activities.

106

RESULTS

Subaqueous Soils and Aquaculture Lease areas

These five coastal ponds currently support a total of 22 total aquaculture leases which occupy 95.3 total acres, or 2% of the surface area coastal ponds when using the

Rhode Island DEM shellfish Regulatory shoreline data available on RIGIS (Table

3.4). Nearly all of the leased acreage uses on-the-bottom production methods. A very small portion of the total lease acreage is used to raise nursery stock within floating upweller systems, or use other floating production methods for grow-out, as these technologies have recently been developed for use in the industry.

Potter Pond currently has the greatest percentage of leased area at 3.3%, followed closely by Point Judith Pond with 3.1%. Ninigret Pond currently has 10 leases that occupy 1.5% of the pond, and Winnapaug and Ninigret ponds have a total of 2% and

1.5% total leased acreage, respectively. Quonochontaug Pond currently has the smallest percentage of leased area at 0.01% with one exploratory lease currently permitted (Table 3.4).

Within Potter and Winnapaug Ponds, 100% of the leased acreage is located on

Psammowassent soils in WFF and WFS soil landscapes. Similarly, 93.4% of the existing lease areas within Ninigret Pond are located on Psammowassent soils in the same soil landscapes, while only 6.6% of the lease areas in Ninigret are on

Sulfiwassent soils located in LB soil landscapes (a portion of these areas identify locations of floating upweller systems). The existing aquaculture leases within Point

Judith Pond are located within Sulfiwassent soils associated with depositional LB soil landscapes, with one lease that is using floating production methods. Therefore, 46%

107

(43.8 ac) of the leased acreage in the coastal ponds is located within Psammowassent soils, 53% (50.5 ac) is located within Sulfiwassent soils, and less than 1% is located in

Haplowassent soils.

Aquaculture Restriction Zone

We compiled spatial data that incorporated submerged aquatic vegetation

(eelgrass), mooring/anchorage areas, docks/piers, navigation corridors, recreational shellfishing areas, and State designated spawner sanctuaries into an “aquaculture restriction zone” (ARZ) for each pond (Table 3.5). Overall, the ARZ occupies nearly

57% or 2,713 acres of the total coastal salt pond region. Therefore, the area potentially available for new leases that may have reduced user conflict is 43% or 2,060 acres of the coastal salt ponds. Recreational shellfishing, existing submerged aquatic vegetation, and navigation activities comprise the largest component of the ARZ, respectively, occupying 82% (2,224.9 ac) of the area. Poor water quality, spawner sanctuaries, and the areas occupied by mooring/anchorage and docks/piers, make up the remaining 18% (488.5 ac) of the ARZ (Table 3.5).

The portion of each pond occupied by the ARZ is variable given the intensity of water dependent uses and amount of eelgras within each pond; however, on average the ARZ reduces the potential aquaculture development area by 66% (Table 3.5).

Potter has the greatest ARZ when compared to the other salts ponds, reducing the potential aquaculture development acreage by 70% within the pond. Winnapaug has the smallest ARZ, reducing the potential aquaculture development acreage by 43%

(Figure 3.2).

108

ARZ and Available Subaqueous Soils

By overlaying the ARZ layer for each pond with existing subaqueous soil mapping data, we were able to compile the portion of each subaqueous soil Great

Group (Psammowassent, Haplowassent, and Sulfiwassent) that would be available for future aquaculture development when considering the total area of each pond (Figure

2, Figure 3.3).

The portion of each soil Great Group available for future aquaculture development is variable depending on the size and configuration of the soil landscapes within each salt pond (Figure 3.2, Figure 3.3). Sulfiwassents comprise just over 30% of the available lease area within Quonochontaug, and Pt. Judith, while these soils comprise 29%, 23%, and 18% of Potter, Winnapaug, and Ninigret, respectively.

Psammowassents comprise as much as 29% of the available lease area in Winnapaug and 24% of Ninigret, as these salt ponds have extensive WFF soil landscapes relative to the other ponds. Only 2% of the available lease area in Potter, and 5% of Pt. Judith is comprised of Psammowassents that is not currently occupied by existing lease areas or identified within the ARZ of each pond. Quonochontaug has the greatest portion of

Haplowassents available (8%) when comparing across the ponds, with 5% in

Winnapaug, and only 1% within Ninigret and Pt. Judith (Figure 3.2).

109

DISCUSSION

The Rhode Island Coastal Resources Management Council (CRMC) regulates the amount of bivalve aquaculture within the Rhode Island salt pond region using the “5%

Rule”. This rule maintains that the total area of each pond leased for bivalve aquaculture cannot exceed 5% of the total surface area of each pond at low-tide

[CRMC Red Book, 300.00(e)(6)]. Therefore, under existing regulations aquaculture expansion within the coastal salt pond region is limited to an additional 143.4 ac of public trust lands. Over the last two years (2014-2015), the Rhode Island Coastal

Resources Management Council (CRMC) received 14 aquaculture lease applications seeking to lease an additional 41 acres of public trust land within the coastal salt pond region. Should these lease areas be approved, the total leased acreage in the ponds would increase nearly 44% to 137 acres, or approximately 2.9% of the coastal ponds that support bivalve aquaculture. This would leave approximately 2% of the remaining coastal salt pond region available for lease development, or approximately100 acres.

The question is how much area is left in the ponds that is not subject to user conflicts is productive for oyster aquaculture?

Results from bivalve aquaculture carrying capacity modeling for the region suggests these ecosystems can sustain a sixty-two-fold increase in bivalve aquaculture production (above 2011 levels) prior to affecting the ecological carrying capacity of the ponds (Byron et al., 2011). Suggesting between 46 – 64% of the surface area of the salt ponds could be under production without significant ecological effects from the aquaculture biomass (Byron, 2011). However, this considers the ponds as a black box,

110

with no subaqueous soils or user conflict spatial data to guide the expansion of oyster aquaculture beyond the 5% rule.

The vast majority of the leases in the salt pond region are located in near sub-tidal waters and utilize on bottom culture with various rack/tray and bag systems as opposed to floating culture methods. This method of oyster culture has proven to be very successful in Rhode Island, as growers are able to produce high value oysters for the half shell market, which are in demand across the region. On bottom grow-out methods, however, require a relatively specific set of criteria for the growers including relatively shallow water depth, and firm (typically sandy) substrate allowing growers to work aquaculture leases on-foot, which reduces overall vessel operation costs.

Without firm substrate, growers are not able to utilize these production methods, and require vessels outfitted with specific dredge or hoisting equipment to recover oysters or grow-out trays/racks from the bottom on deeper aquaculture lease sites, or must switch to floating culture methods, which may be more controversial.

Approximately 30% of the available habitat in Point Judith, Potter and

Quonochontaug Ponds that are outside of ARZs are composed of lagoon bottom landscapes with Sulfiwassent soils (Figure 3.2). The problem with substrates like organic-rich Sulfiwassents is that they have low bearing capacity, are unable to support on bottom rack/bag grow out systems, as the weight of these systems causes subsidence of the gear into the mud and subsequent smothering of oysters. In addition, this soft organic rich mud is easily disturbed from the bottom resulting suspension of silt-sized particles, causing high turbidity and risk of reduced feeding efficiency of bivalves grown in such habitats due to increased ingestion of non-food particles.

111

Growth rates are also soil-type dependent. Still et al., (in preparation) showed that oysters grown on sandy subaqueous soils (Psammowassents and Haplowassents) had above average growth rates when compared oysters grown on silty subaqueous soils

(Sulfiwassents) which had below average growth rates. Coarse textured substrates represent areas of higher flow within estuaries, therefore greater seston flux than fine textured substrates supporting greater growth when compared to fine substrates

(Grizzle and Lutz, 1989; Grizzle and Morin, 1989; Rice, 1992; Rice and Pechenik,

1992). These results indicate that subaqueous soil type is a surrogate for seston flux, and further explains variable growth rates among different soil types within coastal lagoon estuarine systems. Thus, shellfish growth and substrate characteristics is a function of flow and seston flux (food availability) and Psammowassents and

Haplowassents are likely the best locations for oyster aquaculture.

Considering these results, the remaining portion of Psammowassent and

Haplowassent soils within the coastal ponds that area not within ARZs should be considered as priority development areas for aquaculture expansion. For example, the coastal ponds that have extensive washover fan flat and washover fan slope landscapes including Ninigret (386 ac), Winnapaug (139 ac), and Quonochontaug (94 ac) Ponds have the greatest potential for development of these soil landscapes for on-the-bottom aquaculture production (Figure 3.2). Given current regulations that limit aquaculture development to 5% of the surface area of the coastal salt ponds, and number of lease proposals in the past few years, limits to lease development for the region could be reached in the relative near future.

112

Subaqueous soils data can also be applied when considering oyster restoration.

Brown et al., (2013) developed an oyster habitat suitability index model (HSI) for wild populations and oyster restoration activities for the salt pond region. This HSI model indicates that SMB soil landscapes with Haplowassent soils, especially with a bouldery or extremely bouldery surface phase, provides some the of highest quality habitat for oyster restoration activities (Brown et al., 2013). Oysters are typically associated with sand, hard bottom and shell reef substrates and are absent most often from areas of fine substrate (silty and silty clay muds) and areas with high sedimentation rates (Brooks, 1996; Burrell, 1986; Stanley and Sellers, 1986). Firm coarse textured soils with suitable gravel, cobble, boulder, or shell surface features, like many Haplowassents, may be important for oyster recruitment.

Therefore, the Haplowassents which are outside of ARZs should be considered for both development of oyster restoration sites, as well as on-the-bottom aquaculture lease development. Quonochontaug Pond has the greatest extent of these soils (64 ac), followed by Ninigret (23 ac) and Winnapaug (22 ac) (Figure 3.2). A combination of restoration reef and aquaculture development may provide optimum use of these soils as the hard substrate will provide sufficient support for reef structures and limit subsidence.

Recent research on oyster restoration indicates that oyster recruitment, growth, and survival tends to be greater on the upper portions of a “high” constructed reefs where sediment burial, low oxygen conditions, and predation pressures are reduced when compared to “low” constructed reefs and on-bottom restoration efforts (Schulte,

2009; Lenihan, 1999). Firm coarse textured soils with suitable gravel, cobble, boulder,

113

or shell surface features, which are common for Haplowassents, may be important for oyster recruitment, and are more likely to support restoration reefs, than fine textured soils that may have increased sedimentation, and low bearing capacity resulting in greater reef subsidence and burial. Permitting some aquaculture lease areas within these locations, might increase the larval supply for recruitment to restoration reefs and existing hard substrates, as some oyster spawning has been documented on existing lease sites within the ponds, depending on culture practices (S. Brown pers. com).

Caution must be exercised with portions of Haplowassent soils, however, as these mainland beach and mainland cove landscapes can receive submarine groundwater discharge from adjacent upland landscapes (Kroeger and Charette, 2008; Masterson et al., 2007; Scott and Moran 2001; Stachelhaus et al., 2012). The glacial drift deposits within the region can have pH values that range between 4.5 and 5.0 (Rector, 1981).

Thus, groundwater discharging from these acidic soils also have low pH, which may be physiologically stressful for calcifying marine organisms. Still and Stolt (2015) conducted field experiments within Haplowassent soils that received submarine groundwater discharge, and recorded mean pH of 6.57 (±0.09 SE), which was significantly below the mean pH values observed within Psammowassents and

Sulfiwassents. This study also found significantly greater percent shell loss of juvenile oyster shells that were placed in these low pH Haplowassents when compared to

Psammowassents and Sulfiwassents (Still and Stolt., 2015). Therefore, Haplowassents that receive submarine groundwater discharge should be avoided when siting oyster

114

reef restoration sites that use shell substrates as reef materials, as the low pH’s will result in greater dissolution and reduce the persistence of the shell materials.

Several Psammowassent soils including Nagunt and Massapog were identified on the HSI model as low quality oyster restoration habitat due to high energy shifting sands that could lead to premature burial of reef substrates (Brown et al., 2013). In addition, the HSI research indicated that the depositional basin LB Sulfiwassent soils are not suitable habitat for oyster restoration due to high sedimentation rates, low bearing capacity of these soils, and the presence of eelgrass within many of these locations (Brown et al., 2013).

Some recent studies have integrated spatial analysis of competing human uses with ecological carrying capacity modeling to assess user conflict and aquaculture production potential to aid in spatial planning (Bricker et al., 2016; Byron et al., 2011;

Byron et al., 2015, Longdill et al., 2008; Silva et al., 2011). Some of these approaches have drawn criticism, as these tools often require access to comprehensive data to drive complex ecosystem based, and hydrodynamic modeling (Bricker et al., 2016).

Additionally, many regions in the U.S. do not have substrate maps with enough detail to inform decisions about substrate characteristics within small estuaries, like coastal lagoons. The coarse resolution surface sediment mapping that may exist are likely not suitable for within estuarine habitat scale management decisions.

115

Summary and Conclusions

Coastal mangers and researchers have expressed the need for tools that balance the growth of bivalve aquaculture with many stakeholders and other coastal zone user groups to avoid user conflicts. We developed this decision support tool with high- resolution habitat and landscape scale subaqueous soils maps with oyster growth rate data, and conflicting use information from within these same systems, to investigate the utility of using these standardized subaqueous soil maps as a tool for spatial planning. Our research shows that this approach can provide the needed resolution for aquaculture lease development, as well as provide useful information for restoration site selection within shallow lagoon estuarine systems.

With this approach, we demonstrated that conflicting uses including eelgrass, recreational shellfishing, recreational boating and navigation, and poor water quality contribute to reducing areas available for aquaculture development within the salt pond region. Of the areas that are available for potential aquaculture development, a majority of the habitat in three of the five coastal salt ponds is comprised of

Sulfiwassent soils, which are not well suited for on-the-bottom aquaculture production. The remaining acreage available for aquaculture lease development is limited to approximately 143 acres due to current regulations limiting aquaculture to

5% of the surface area of the water bodies (“5% rule”). The Psammowassents and

Haplowassents, which are hard sandy substrates, should be prioritized for aquaculture development as these soils have increased growth rates compared to Sulfiwassents, when using on-the-bottom culture methods.

116

Although these studies are being tested and applied in southern New England estuaries, our intent was to develop approaches to the science that can be applied and tested in additional coastal regions throughout the U.S. The Rhode Island salt pond region as well as other coastal lagoon systems along the Atlantic coast of the U.S. including Massachusetts, Rhode Island, Connecticut, New York, New Jersey,

Delaware, Maryland, Virginia, and North Carolina among others, support a multitude of water dependent uses, including bivalve aquaculture. As the extent of subaqueous soil survey continues to expand with extensive mapping efforts underway by the

NRCS and LPSES in Connecticut and New Jersey, and existing maps available

Maryland, Delaware, and Rhode Island, subaqueous soils data will increasingly be available as a planning tool for coastal managers supporting the growth of the bivalve aquaculture industry, while managing user conflict.

117

ACKNOWLEDGMENTS

This project was funded by the Rhode Island Agricultural Experiment Station through funds projected for Multistate Project NE-1038. We would like to thank The

Rhode Island Environmental Data Center and the Coastal Institute at the University of

Rhode Island for support and access to GIS resources used for this project. This paper is a contribution of the Rhode Island Agricultural Experiment Station (no. 5416).

118

CITATIONS

Alleway, H.K., and S.D. Connell. 2015. Loss of an ecological baseline through the eradication of oyster reefs from coastal ecosystems and human memory. Conserv. Biol. 29:795-804.

Beck, M.W., R.D. Brumbaugh, L. Airoldi, A. Carranza, L.D. Coen, C. Crawford, O. Defeo, G.J. Edgar, B. Hancock, M.C. Kay, H.S. Lenihan, M.W. Luckenbach, C.L. Toropova, G. Zhang, and X. Guo. 2011. Oyster reefs at risk and recommendations for conservation, restoration, and management. BioScience. 61:107-116.

Beutel, D. 2014. Aquaculture in Rhode Island 2014 Annual Status Report. Rhode Island Coastal Resources Management Council. 10p.

Boothroyd, J.C., N.E. Friedrich, and S.R. McGinn. 1985. Geology of microtidal coastal lagoons: Rhode Island. Mar. Geol. 63:35–76.

Bowen J.L., and I. Valiela, 2001. The ecological effects of urbanization of coastal watersheds: historical increases in nitrogen loads and eutrophication of Waquoit Bay estuaries. Can. J. Fish. Aquat. Sci. 58:1489-1500.

Bradley, M.P., and M.H. Stolt. 2003. Subaqueous soil–landscape relationships in a Rhode Island estuary. Soil Sci. Soc. Am. J. 67:1487–1495.

Bradley, M.P., and M.H. Stolt. 2006. Landscape-level seagrass-sediment relations in a coastal lagoon. Aquat. Bot. 84:121–128.

Bricker, S.B., T.L. Getchis, C.B. Chadwick, C.M. Rose, and J.M. Rose. 2016. Integration of ecosystem-based models into an existing interactive web-based tool for improved aquaculture decision-making. Aquaculture. 453:135-146.

Brooks, W. K. 1996. The oyster. Johns Hopkins University Press. 230p.

Brown, S.B., K. Ruddock, J. O’Brien, G. Casabona and J.A. Bachand. 2013. Environmental Quality Incentives Program (EQIP)—Oyster Restoration: Final Programmatic Report. USDA Natural Resource Conservation Service, Conservation Technical Assistance Program.

Burrell, V.G. 1986. Species profiles: life histories and environmental requirements of coastal fishes and invertebrates (South Atlantic) American oyster. U.S. Fish and Wildl. Serv. Biol. Rep. 82 (11.57). U.S. Army Corps of Engineers TR EL-82-4.

Byron, C., Link, J., Costa-Pierce, B., and D. Bengtson. 2011. Modeling ecological carrying capacity of shellfish aquaculture in highly flushed temperate lagoons. Aquaculture. 314:87-99.

119

Byron, C.J., D. Jin, and T.M. Dalton. 2015. An Integrated ecological–economic modeling framework for the sustainable management of oyster farming. Aquaculture. 447:15-22.

Coastal Resources Management Council, 2008. CRMC working group on aquaculture regulations, report on the deliberations of the group and suggestions to guide future development.88 p. (http://www.crmc.ri.gov/aquaculture.html#riag) [accessed 3/1/2016].

Clements, J. C., and T. Chopin. 2016. Ocean acidification and marine aquaculture in North America: potential impacts and mitigation strategies. Reviews in Aquaculture 0:1-16.

Cochrane, K., C. De Young, D. Soto, and T. Bahri, 2009. Climate change implications for fisheries and aquaculture. FAO Fisheries and aquaculture technical paper. 530. 212p.

Crossett, K.M., T.J. Culliton, P.C. Wiley, and T.R. Goodspeed. 2004. Population trends along the coastal United States: 1980-2008. Coastal Trends Report Series. National Oceanic and Atmospheric Administration, NOAA’s National Ocean Service, Special Projects. http://coastalsocioeconomics.noaa.gov/

Davies, J.L. 1964. A morphogenetic approach to world shorelines. Geomorphol. 8:127–142.

Deegan, L.A., A. Wright, S.G. Ayvazian, J.T. Finn, H. Golden, R.R. Merson, and J. Harrison. 2002. Nitrogen loading alters seagrass ecosystem structure and support of higher trophic levels. Aquat. Conserv. 12:193-212.

East Coast Shellfish Growers Association. 2014. Atlantic coastal state shellfish farming lease or permit regulations. (http://www.ecsga.org/Pages/Regulations/StateRegulations2014.pdf) [accessed 3/17/2016].

Ernst, Laura.M., L.K. Miguel, and J. Willis. 1999. Rhode Island's Salt Pond Region: A Special Area Management Plan (Maschaug to Point Judith Ponds) for the Salt Pond Watersheds in the Towns of Westerly, Charlestown, South Kingstown and Narragansett, April 12, 1999. Rhode Island Coastal Resources Management Council.

ESRI, 2014. ArcGIS Desktop: Release 10.2.2 Redlands, CA: Environmental Systems Research Institute.

120

Grizzle, R. E., and R.A. Lutz. 1989. A statistical model relating horizontal seston fluxes and bottom sediment characteristics to growth of Mercenaria mercenaria. Mar. Biol. 102:95-105.

Grizzle, R.E., and P.J. Morin. 1989. Effects of tidal currents, seston, and bottom sediments on growth of Mercenaria mercenaria: results of a field experiment. Mar. Biol. 102:85-93.

Halpern, B.S., S. Walbridge, K.A. Selkoe, C.V. Kappel, F. Micheli, C. D'Agrosa, J.F. Bruno, K.S. Casey, C. Ebert, H.E. Fox, and R. Fujita. 2008. A global map of human impact on marine ecosystems. Science. 319:948-952.

Jackson, J.B., M.X. Kirby, W.H. Berger, K.A. Bjorndal, L.W. Botsford, B.J. Bourque, and T.P. Hughes. 2001. Historical overfishing and the recent collapse of coastal ecosystems. Science. 293:629-637.

Jespersen, J.L., and L.J. Osher. 2007. Carbon storage in the soils of a mesotidal Gulf of Maine estuary. Soil Sci. Soc. Am. J. 71:372–379.

Knapp, G. and Rubino, M.C., 2016. The Political Economics of Marine Aquaculture in the United States. Rev. Fish. Sci. Aqua. 24:213-229.

Kroeger, K.D. and M.A. Charette. 2008. Nitrogen biogeochemistry of submarine groundwater discharge. Limnol. Oceanogr. 53:1025–1039

Lenihan, H.S. 1999. Physical-biological coupling on oyster reefs: how habitat structure influences individual performance. Ecol. Mon. 19:251-275.

Longdill, P.C., T.R. Healy, and K.P. Black. 2008. An integrated GIS approach for sustainable aquaculture management area site selection. Ocean Coast Manage. 51:612-624.

Mackenzie, C.L. 2007. Causes underlying the historical decline in eastern oyster (Crassostrea virginica Gmelin, 1791) landings. J. Shellfish Res. 26(4):927-938.

121

Masterson, J.P., J.R. Sorenson, J.R. Stone, S.B. Moran, and A. Hougham. 2007 Hydrogeology and simulated ground-water flow in the Salt Pond region of southern Rhode Island: U.S. Geological Survey, Scientific Investigations Report 2006–5271. 56 p.

Millar, C.M., A.A.O. Aduomih, B. Still, and M.H. Stolt. 2015. Estuarine subaqueous soil organic carbon accounting: Sequestration and storage. Soil Sci. Soc. Am. J. 79:389-397.

National Ocean Policy Implementation Plan. 2013. National Ocean Council, Washington DC available at (https://www.whitehouse.gov/administration/eop/oceans/implementationplan) [accessed 3/17/2016].

NOAA, National Oceanic Atmospheric Administration, National Shellfish Initiative. 2011. http://www.nmfs.noaa.gov/aquaculture/policy/shellfish_initiative_homepage.html . [Accessed 3/7/2016].

Nixon S.W. 1993. Nutrients and coastal waters, too much of a good thing? Oceanus. 35:38-47.

Nixon, S. W. 1995. Coastal marine eutrophication: A definition, social causes, and future concerns. Ophelia. 41:199-219.

Nixon S.W., B. Buckley, S. Granger, and J. Bintz. 2001. Responses of very shallow marine systems to nutrient enrichment. Hum. Ecol. Risk Assess. 7:1457-1481.

Pacific Coast Shellfish Growers Association. Washington Shellfish Initiate. 2011. (http://pcsga.org/wprs/wp-content/uploads/2013/04/Washington-Shellfish- Initiative.pdf) [Accessed 3/17/2016].

Payne, M.K. 2007. Landscape-Level Assessment of Subaqueous Soil and Water Quality in Shallow Embayments in Southern New England. M. S. Thesis. Univ. of Rhode Island, Kingston, RI.

Rector, D.D. 1981. Soil survey of Rhode Island. US Gov. Print. Office, Washington, DC.

122

RIGIS, 2016. Rhode Island Geographic Information System (RIGIS) Data Distribution System. URL: http://www.edc.uri.edu/rigis, Environmental Data Center, University of Rhode Island, Kingston, Rhode Island [accessed 3/1/2016].

Rhode Island Department of Environmental Management. 2015. Notice of polluted shellfish grounds. RIDEM Office of Water Resources. 45 p.

Rhode Island Shellfish Management Plan (SMP). 2014. http://www.rismp.org/. [accessed 3/1/2016].

Rothschild, B. J., Ault, J. S., Goulletquer, P., and M. Heral. 1994. Decline of the Chesapeake Bay oyster population: a century of habitat destruction and overfishing. Mar Eco Prog Ser. 111:29-39.

Salisbury, A.R. 2010. Developing subaqueous soil interpretations for Rhode Island estuaries. M.S. thesis. Univ. of Rhode Island, Kingston.

Schulte, D.M., R. P. Burke, and R. N. Lipcius (2009). Unprecedented restoration of a native oyster metapopulation. Science 325:1124-1128.

Scott, M.K., and S.B. Moran. 2001. Ground water input to coastal salt ponds of southern Rhode Island estimated using 226 Ra as a tracer. J. Environ. Radioactiv. 54:163-174.

Seitz, R.D., 2014. Value of coastal habitats for exploited species: introduction to a theme set of articles. ICES J. Mar. Sci.71:636-637.

Short, F.T., D.M. Burdick, S. Granger, and S.W. Nixon. 1996. Long-term decline in eelgrass Zostera marina L., linked to increased housing development. Seagrass Biology Proceedings of an International Workshop. Eds J. Kuo, R.C. Phillips, D.I. Walker, and H. Kirkman. Pp.291-298.

Silva, C., J.G. Ferreira, S.B. Bricker, T.A. DelValls, M.L. Martín-Díaz, and E Yáñez. 2011. Site selection for shellfish aquaculture by means of GIS and farm-scale models, with an emphasis on data-poor environments. Aquaculture. 318:444-457.

Soil Survey Division Staff. 1993. Soil survey manual. United States Department of Agriculture. 315p.

Soil Survey Staff. 1999. Soil Taxonomy. U.S. Dept. Agric. Handbook No. 436, second ed. US Govt. Printing Office, Washington, DC. 886p.

Soil Survey Staff. 2009. Soil Survey Field and Laboratory Methods Manual. Soil Survey Investigations Report No. 51, Version 1.0. R. Burt (ed.). U.S. Department of Agriculture, Natural Resources Conservation Service. 487p.

123

Soil Survey Staff. 2010. Keys to Soil Taxonomy, 11th ed. USDA-Natural Resources Conservation Service, Washington, DC. 372p.

Soil Survey Staff. 2016. Natural Resources Conservation Service, United States Department of Agriculture. Web Soil Survey. Available online at http://websoilsurvey.nrcs.usda.gov/. Accessed [3/17/2016].

Stachelhaus, S. L., S.B. Moran, and R.P. Kelly. 2012. An evaluation of the efficacy of radium isotopes as tracers of submarine groundwater discharge to southern Rhode Island's coastal ponds. Mar. Chem.130:49-61.

Stanley, J.G., and M.A. Sellers. 1986. Species Profiles: Life histories and environmental requirements of coastal fishes and invertebrates (Mid-Atlantic) American Oyster. U.S. Fish Wildl. Serv. Biol. Rep. 82(11.65) U.S. Army Corps of Engineers, TR EL-82-4.

Still B.M. and M.H. Stolt. 2015. Subaqueous soils and coastal acidification: A hydropedology perspective with implications for calcifying organisms. Soil Sci. Soc. Am. J. 79:407–416.

Stolt, M., M. Bradley, J. Turenne, M. Payne, E. Scherer, G. Cicchetti, et al. 2011. Mapping shallow coastal ecosystems: A case study of a Rhode Island lagoon. J. Coast. Res. 27:1–15.

Stolt, M.H., and M.C. Rabenhorst. 2011. Subaqueous soils. In: P.M. Huang et al., editors, Handbook of soil science. 2nd ed. CRC Press, Boca Raton, FL. p. 36-1– 36-14.

USDA 2015 State Agriculture Overview. Rhode Island (http://www.nass.usda.gov/Quick_Stats/Ag_Overview/stateOverview.php?state= RHODE%20ISLAND). (accessed 3/17/2016).

USDA Census of Aquaculture.2013. Vol 3. Special Studies Part 2. 2014. 2014. AC- 12-SS-2. United States Department of Agriculture, National Agricultural Statistics Service. 98.p

124

TABLES

125

Table 3.1. Current bivalve aquaculture production value and lease acreage within the New England region. Acres Leased Farms Production Production State Primary Species (km2) (leases) Value (millions) per acre Eastern Oyster ME1 600(2.43) 74 $4.0 $6,667.00 Blue Mussel Eastern oyster, NH2 186(0.75) 22 $2.5 $13,500 hard clam, blue mussel Eastern oyster, MA3 1,030(4.17) 378 $10.2 $9,903.00 hard clam, Eastern oyster, RI4 206.2(0.83) 55 $5.2 $25,242.00 hard clam, blue mussel

CT5 73,091(295.79) 998 $25.4 $410.00 Eastern oyster

1 http://www.maine.gov/dmr/aquaculture. Based on 2014 data 2http://www.granit.unh.edu. 2014 GIS data of aquaculture farms. Production value data for NH was not available Values estimated based on average production/acre from ME, MA, and RI 3http://extension.umass.edu/aquaculture/publications-and-resources. Based on 2010 data 4http://www.crmc.ri.gov/aquaculture.html. Based on 2014 data 5http://www.ct.gov/doag. Based on the most current production information

126

Data Layers Source Description

Generated 25 ft buffer established around all doc/pier Docks/Piers (this study) structures within each pond

Generated 150 ft navigation corridor established for Navigation Centerline (this study) heavily used areas of each pond.

Generated Mooring/anchorage areas identified by Mooring/Anchorage (this study) digitizing from summer period aerial imagery

Combined footprint of submerged aquatic Potential SAV areas RIGIS vegetation from 2009 and 2012 collected by the RI Eelgrass Mapping Taskforce

Recreational shellfishing areas identified on Recreational Shellfishing RI SMP the RI Shellfish Management Plan user maps

Shellfish Spawner Sanctuaries designated by RIDEM Spawner Sanctuary RIDEM RI Department of Environmental Management

RI Department of Environmental Management RIDEM Shellfish RIGIS regulatory shellfish areas. Data includes areas prohibited from the harvesting of shellfish

127

Table 3.3. The total acreage and relative percent of subaqueous soil Great Groups within the coastal ponds. The subaqueous soils data are aggregated at the Great Group level as Haplowassents, Psammowassents, and Sulfiwassents. The total percent data represents the percent of the total acreage occupied by each of the subaqueous soil Great Groups.

Haplowassents Psammowassents Sulfiwassents Coastal Ponds (area in acres) (ac) (%) (ac) (%) (ac) (%)

Pt. Judith 18.2 1 436.8 28 1,125.5 71 (1,580)

Potter - - 122.7 34 237.6 66 (360)

Ninigret 67.8 4 589.2 37 944.2 59 (1601)

Quonochontaug 105.8 14 213.1 28 440.6 58 (760)

Winnapaug 52.8 11 235.2 50 184.2 39 (472)

Total 244.6 5 1,597.0 33 2,932.1 61 (4,773)

128

Table 3.4. Existing aquaculture lease acreage in the coastal salt pond region.

Coastal Ponds Existing Lease Area Existing % of # of Existing (area in ac) (ac) Pond Area Leases

Pt. Judith 49 3.1 7 (1,580)

Potters 11.8 3.3 1 (360)

Ninigret 24.3 1.5 10 (1,601)

Quonochontaug 0.75 0.1 1 (760)

Winnapaug 9.4 1.99 3 (472)

Total 95.3 2.0 22 (4,773)

129

Incompatible Uses for Aquaculture Development Total Moorings Recreational Spawner Total Available Coastal Ponds Eelgrass1 Docks/ Navigation3 Prohibited5 Shellfishing4 Sanctuary6 ARZ7 Lease Area Piers2 (%)

Pt. Judith 135.2 154.3 183.2 324.1 425.7 - 988.9 591.5 (37)

Potter 142.2 12.8 68 92.5 26.3 11 250.3 109.9 (31)

Ninigret 353.9 33.4 136.9 329.5 - 177.6 908.8 692.3 (43)

Quonochontaug 117 46.9 98.6 84.2 - 76.3 362.6 396.9 (52)

130 Winnapaug - 4 71.7 87.9 6.2 40 202.8 269.3 (57)

Total Area 748.3 251.4 558.4 918.2 458.2 304.9 2713.4 2059.9 (43)

% Area 16 5 12 19 10 6 57 43 All numbers are reported in acres or percent when identified (%) 1 eelgrass includes area occupied by eelgrass during the two most recent surveys (2009 2012) conducted by state eelgrass monitoring network 2 Mooring data digitized from summer season aerial photos. Docks piers represents 25ft buffer from edge of dock/pier. 3 Navigation area represents a 150ft navigation corridor through the ponds including large coves that include docs/piers 4 Recreational shellfish area data adapted from the Rhode Island Shellfish Management Plan use maps 5. Prohibited areas adapted from latest RI DEM shellfish closure information 6 Spawner sanctuary areas adapted from RI DEM shellfish regulations 7 Total ARZ represents a merged polygon that includes all incompatible use areas. Individual use areas will not sum to reported total ARZ due to overlapping incompatible use

FIGURES

131

Figure 3.1. Site map of southern Rhode Island coastal salt pond region. Labeled coastal salt ponds that currently support bivalve aquaculture leases.

132

Figure 3.2. The pie charts represent the acreage and percent of each coastal pond that is occupied by the aquaculture restriction zone (ARZ) and potentially available areas open to bivalve aquaculture development. The coastal ponds with the largest washover fan landscapes Ninigret, Winnapaug and Quonochontaug ponds respectively have the greatest acreage of Psammowassents available for bivalve aquaculture development.

133

Figure 3.3. Map depicts subaqueous soils classified at the great group level within the coastal salt ponds that support aquaculture. Areas likely to not support the development of bivalve aquaculture due to incompatible use (eelgrass, navigation/moorings, shellfish spawner sanctuaries, recreational shellfishing, and areas prohibited from shellfish harvest due to poor water quality) are depicted as an aquaculture restriction zone (ARZ) within each pond. The available Psammowassent soils would best support on-the-bottom production methods, while the Sulfiwassents would best support floating culture production. The Haplowassents could also support aquaculture development, however a portion of these soils have also been identified as prime areas for oyster restoration activities by habitat suitability index (HIS) modeling. The white polygons within each pond represent existing bivalve aquaculture lease areas.

134